Instructions to use hermitdave/K2-Horizon-7B-Uno-oQ4e with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hermitdave/K2-Horizon-7B-Uno-oQ4e with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("hermitdave/K2-Horizon-7B-Uno-oQ4e")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use hermitdave/K2-Horizon-7B-Uno-oQ4e with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "hermitdave/K2-Horizon-7B-Uno-oQ4e"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "hermitdave/K2-Horizon-7B-Uno-oQ4e"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use hermitdave/K2-Horizon-7B-Uno-oQ4e with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "hermitdave/K2-Horizon-7B-Uno-oQ4e"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "hermitdave/K2-Horizon-7B-Uno-oQ4e"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "hermitdave/K2-Horizon-7B-Uno-oQ4e",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use hermitdave/K2-Horizon-7B-Uno-oQ4e with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "hermitdave/K2-Horizon-7B-Uno-oQ4e"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default hermitdave/K2-Horizon-7B-Uno-oQ4e

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use hermitdave/K2-Horizon-7B-Uno-oQ4e with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "hermitdave/K2-Horizon-7B-Uno-oQ4e"

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "hermitdave/K2-Horizon-7B-Uno-oQ4e" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

hermitdave commited on 8 days ago

Commit

bb03345

verified ·

1 Parent(s): 5f2c1ae

Add files using upload-large-folder tool

Browse files

Files changed (9) hide show

.gitattributes +1 -0
README.md +66 -0
config.json +712 -0
configuration_k2_horizon.py +96 -0
model-00001-of-PLACEHOLDER.safetensors +3 -0
modeling_k2_horizon.py +1116 -0
oq_imatrix_report.json +364 -0
tokenizer.json +3 -0
tokenizer_config.json +10 -0

.gitattributes CHANGED Viewed

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md ADDED Viewed

	@@ -0,0 +1,66 @@

+---
+base_model: IFM/K2-Horizon-7B-Uno
+tags:
+  - mlx
+  - apple-silicon
+  - text-generation
+  - uno
+  - oQ
+  - oQ4e
+license: apache-2.0
+---
+# K2-Horizon-7B-Uno oQ4e
+oQ4e (imatrix-enhanced mixed-precision ~4.5 BPW) quantization of the merged [IFM/K2-Horizon-7B-Uno](https://huggingface.co/IFM/K2-Horizon-7B-Uno) model — a diffusion-augmented LLM based on K2-Horizon-7B. The LoRA adapter is baked into the base weights, so it runs as a standard autoregressive model.
+**Upstream model:** [IFM/K2-Horizon-7B-Uno](https://huggingface.co/IFM/K2-Horizon-7B-Uno) by Institute of Foundation Models, released under Apache 2.0.
+**Conversion:** Merged and quantized to MLX format using [Hermes Agent](https://hermes-agent.nousresearch.com) with `oMLX`.
+## Quickstart
+```bash
+pip install -U mlx-lm
+python3 -m mlx_lm.generate \
+  --model hermitdave/K2-Horizon-7B-Uno-oQ4e \
+  --prompt "Explain step by step." \
+  --max-tokens 512 --temp 1.0 --top-p 0.95
+```
+## Reasoning
+K2-Horizon-7B is a reasoning model. Always use `reasoning_effort="high"`:
+```python
+from openai import OpenAI
+client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY")
+response = client.chat.completions.create(
+    model="hermitdave/K2-Horizon-7B-Uno-oQ4e",
+    messages=[{"role": "user", "content": "Explain step by step."}],
+    extra_body={"chat_template_kwargs": {"reasoning_effort": "high"}},
+)
+print("Reasoning:", getattr(response.choices[0].message, "reasoning_content", None))
+print("Answer:", response.choices[0].message.content)
+```
+## oMLX Patch
+K2-Horizon requires oMLX v0.6.4+ with the [K2-Horizon support patch](https://github.com/jundot/omlx/pull/3441).
+## Citation
+```bibtex
+@misc{k2_horizon_7b_uno,
+  title        = {K2-Horizon-7B-Uno},
+  author       = {Institute of Foundation Models},
+  year         = {2026},
+  howpublished = {\url{https://huggingface.co/IFM/K2-Horizon-7B-Uno}},
+}
+```
+## License
+Apache 2.0 (same as upstream).

config.json ADDED Viewed

	@@ -0,0 +1,712 @@

+{
+  "architectures": [
+    "K2HorizonForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attention_gate_func": null,
+  "auto_map": {
+    "AutoConfig": "configuration_k2_horizon.K2HorizonConfig",
+    "AutoModel": "modeling_k2_horizon.K2HorizonModel",
+    "AutoModelForCausalLM": "modeling_k2_horizon.K2HorizonForCausalLM"
+  },
+  "bos_token_id": 0,
+  "decoder_sparse_step": 1,
+  "dtype": "bfloat16",
+  "eos_token_id": 1,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 12288,
+  "layernorm_num_groups": 4,
+  "max_position_embeddings": 524288,
+  "mlp_only_layers": [
+    0,
+    1,
+    2,
+    3,
+    4,
+    5,
+    6,
+    7,
+    8,
+    9,
+    10,
+    11,
+    12,
+    13,
+    14,
+    15,
+    16,
+    17,
+    18,
+    19,
+    20,
+    21,
+    22,
+    23,
+    24,
+    25,
+    26,
+    27,
+    28,
+    29,
+    30,
+    31,
+    32,
+    33,
+    34,
+    35
+  ],
+  "model_type": "k2_horizon",
+  "moe_gate_bias": false,
+  "moe_intermediate_size": 0,
+  "mova_num_experts": 0,
+  "mova_num_experts_per_tok": 0,
+  "norm_topk_prob": true,
+  "num_attention_heads": 32,
+  "num_experts": 0,
+  "num_experts_per_tok": 0,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 8,
+  "num_shared_experts": 0,
+  "output_router_logits": false,
+  "pad_token_id": null,
+  "query_key_norm": false,
+  "rms_norm_eps": 1e-06,
+  "rope_head_dim": 128,
+  "rope_parameters": {
+    "rope_theta": 10000000.0,
+    "rope_type": "default"
+  },
+  "router_aux_loss_coef": 0.001,
+  "router_scaling_factor": 1.0,
+  "router_score_func": "sigmoid",
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "5.13.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 250624,
+  "quantization": {
+    "group_size": 64,
+    "bits": 4,
+    "mode": "affine",
+    "model.layers.0.mlp.down_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.0.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.0.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.0.self_attn.q_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.0.self_attn.v_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.mlp.down_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.self_attn.q_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.self_attn.v_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.2.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.2.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.3.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.3.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.4.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.10.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.10.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.4.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.5.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.5.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.6.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.6.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.7.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.7.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.8.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.8.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.9.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.9.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.11.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.11.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.12.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.12.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.13.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.13.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.14.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.14.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.15.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.15.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.16.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.16.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.17.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.17.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.18.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.18.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.19.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.19.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.20.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.20.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.21.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.21.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.22.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.22.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.23.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.23.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.23.self_attn.q_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.23.self_attn.v_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.24.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.24.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.24.self_attn.q_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.24.self_attn.v_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.25.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    }
+  },
+  "quantization_config": {
+    "group_size": 64,
+    "bits": 4,
+    "mode": "affine",
+    "model.layers.0.mlp.down_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.0.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.0.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.0.self_attn.q_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.0.self_attn.v_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.mlp.down_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.self_attn.q_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.1.self_attn.v_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.2.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.2.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.3.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.3.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.4.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.10.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.10.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.4.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.5.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.5.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.6.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.6.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.7.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.7.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.8.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.8.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.9.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.9.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.11.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.11.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.12.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.12.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.13.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.13.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.14.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.14.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.15.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.15.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.16.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.16.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.17.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.17.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.18.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.18.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.19.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.19.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.20.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.20.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.21.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.21.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.22.mlp.down_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.22.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.23.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.23.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.23.self_attn.q_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.23.self_attn.v_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.24.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.24.self_attn.o_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.24.self_attn.q_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.24.self_attn.v_proj": {
+      "bits": 6,
+      "group_size": 64,
+      "mode": "affine"
+    },
+    "model.layers.25.self_attn.k_proj": {
+      "bits": 5,
+      "group_size": 64,
+      "mode": "affine"
+    }
+  }
+}

configuration_k2_horizon.py ADDED Viewed

	@@ -0,0 +1,96 @@

+# Copyright 2024 The Qwen team, Alibaba Group and the HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""K2Horizon model configuration"""
+from huggingface_hub.dataclasses import strict
+from transformers.configuration_utils import PreTrainedConfig
+from transformers.modeling_rope_utils import RopeParameters
+@strict
+class K2HorizonConfig(PreTrainedConfig):
+    r"""
+    decoder_sparse_step (`int`, *optional*, defaults to 1):
+        The frequency of the MoE layer.
+    mlp_only_layers (`list[int]`, *optional*, defaults to `[]`):
+        Indicate which layers use K2HorizonMLP rather than K2HorizonSparseMoeBlock
+        The list contains layer index, from 0 to num_layers-1 if we have num_layers layers
+        If `mlp_only_layers` is empty, `decoder_sparse_step` is used to determine the sparsity.
+    ```python
+    >>> from transformers import K2HorizonModel, K2HorizonConfig
+    >>> # Initializing a K2Horizon style configuration
+    >>> configuration = K2HorizonConfig()
+    >>> model = K2HorizonModel(configuration)
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```
+    """
+    model_type = "k2_horizon"
+    keys_to_ignore_at_inference = ["past_key_values"]
+    vocab_size: int = 151936
+    hidden_size: int = 2048
+    intermediate_size: int = 6144
+    num_hidden_layers: int = 24
+    num_attention_heads: int = 32
+    num_key_value_heads: int = 4
+    hidden_act: str = "silu"
+    max_position_embeddings: int = 32768
+    initializer_range: float = 0.02
+    rms_norm_eps: float = 1e-6
+    use_cache: bool = True
+    tie_word_embeddings: bool = False
+    rope_parameters: RopeParameters | dict | None = None
+    attention_bias: bool = False
+    use_sliding_window: bool = False
+    sliding_window: int | None = 4096
+    attention_dropout: float | int = 0.0
+    decoder_sparse_step: int = 1
+    moe_intermediate_size: int = 768
+    num_experts_per_tok: int = 8
+    num_experts: int = 128
+    norm_topk_prob: bool = False
+    output_router_logits: bool = False
+    router_aux_loss_coef: float = 0.001
+    mlp_only_layers: list[int] | None = None
+    pad_token_id: int | None = None
+    bos_token_id: int | None = None
+    eos_token_id: int | list[int] | None = None
+    head_dim: int = 128
+    query_key_norm: bool = True
+    moe_gate_bias: bool = False
+    layernorm_num_groups: int = 1
+    num_shared_experts: int = 0
+    router_score_func: str = "softmax"
+    router_scaling_factor: float | None = 1.0
+    rope_head_dim: int | None = None
+    attention_gate_func: str | None = None
+    mova_num_experts: int = 0
+    mova_num_experts_per_tok: int = 0
+    def __post_init__(self, **kwargs):
+        self.sliding_window = self.sliding_window if self.use_sliding_window else None
+        self.mlp_only_layers = [] if self.mlp_only_layers is None else self.mlp_only_layers
+        if self.router_scaling_factor is None:
+            self.router_scaling_factor = 1.0
+        super().__post_init__(**kwargs)
+__all__ = ["K2HorizonConfig"]

model-00001-of-PLACEHOLDER.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:39886a6724d9f346dbeb1b36c0f7349cf2f46a4f732ed658c9954ae16907484f
+size 5287482670

modeling_k2_horizon.py ADDED Viewed

	@@ -0,0 +1,1116 @@

+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+#           This file was automatically generated from src/transformers/models/qwen3_moe/modular_qwen3_moe.py.
+#               Do NOT edit this file manually as any edits will be overwritten by the generation of
+#             the file from the modular. If any change should be done, please apply the change to the
+#                          modular_qwen3_moe.py file directly. One of our CI enforces this.
+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+# coding=utf-8
+# Copyright 2025 The Qwen team, Alibaba Group and the HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from typing import Callable, Optional, Union
+import math
+import torch
+import torch.nn.functional as F
+from torch import nn
+from transformers.activations import ACT2FN
+from transformers.cache_utils import Cache, DynamicCache
+from transformers.generation import GenerationMixin
+from transformers.masking_utils import create_causal_mask, create_sliding_window_causal_mask
+from transformers.modeling_flash_attention_utils import FlashAttentionKwargs
+from transformers.modeling_layers import (
+    GenericForQuestionAnswering,
+    GenericForSequenceClassification,
+    GenericForTokenClassification,
+    GradientCheckpointingLayer,
+)
+from transformers.modeling_outputs import MoeCausalLMOutputWithPast, MoeModelOutputWithPast
+from transformers.modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update
+from transformers.modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
+from transformers.processing_utils import Unpack
+from transformers.utils import TransformersKwargs, auto_docstring, can_return_tuple
+from transformers.utils.generic import maybe_autocast
+from transformers.utils.deprecation import deprecate_kwarg
+from transformers.utils.output_capturing import OutputRecorder
+from .configuration_k2_horizon import K2HorizonConfig
+def rotate_half(x):
+    """Rotates half the hidden dims of the input."""
+    x1 = x[..., : x.shape[-1] // 2]
+    x2 = x[..., x.shape[-1] // 2:]
+    return torch.cat((-x2, x1), dim=-1)
+def apply_rotary_pos_emb(q, k, cos, sin, position_ids=None, unsqueeze_dim=1):
+    """Applies Rotary Position Embedding to the query and key tensors.
+    Args:
+        q (`torch.Tensor`): The query tensor.
+        k (`torch.Tensor`): The key tensor.
+        cos (`torch.Tensor`): The cosine part of the rotary embedding.
+        sin (`torch.Tensor`): The sine part of the rotary embedding.
+        position_ids (`torch.Tensor`, *optional*):
+            Deprecated and unused.
+        unsqueeze_dim (`int`, *optional*, defaults to 1):
+            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
+            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
+            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
+            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
+            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
+            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
+    Returns:
+        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
+    """
+    cos = cos.unsqueeze(unsqueeze_dim)
+    sin = sin.unsqueeze(unsqueeze_dim)
+    q_embed = (q * cos) + (rotate_half(q) * sin)
+    k_embed = (k * cos) + (rotate_half(k) * sin)
+    return q_embed, k_embed
+def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
+    """
+    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
+    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
+    """
+    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
+    if n_rep == 1:
+        return hidden_states
+    hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
+    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
+def split_to_interleaved(x):
+    # Split halves:  x0 x1 x2 x3 ... y0 y1 y2 y3 ...
+    # Interleaved:   x0 y0 x1 y1 x2 y2 x3 y3 ...
+    return x.reshape(*x.shape[:-1], 2, -1).transpose(-1, -2).reshape(*x.shape[:-1], -1)
+def interleaved_to_split(x):
+    # Interleaved:   x0 y0 x1 y1 x2 y2 x3 y3 ...
+    # Split halves:  x0 x1 x2 x3 ... y0 y1 y2 y3 ...
+    return x.reshape(*x.shape[:-1], -1, 2).transpose(-1, -2).reshape(*x.shape[:-1], -1)
+def eager_attention_forward(
+        module: nn.Module,
+        query: torch.Tensor,
+        key: torch.Tensor,
+        value: torch.Tensor,
+        attention_mask: Optional[torch.Tensor],
+        scaling: float,
+        dropout: float = 0.0,
+        **kwargs: Unpack[TransformersKwargs],
+):
+    key_states = repeat_kv(key, module.num_key_value_groups)
+    value_states = repeat_kv(value, module.num_key_value_groups)
+    attn_weights = torch.matmul(query, key_states.transpose(2, 3)) * scaling
+    if attention_mask is not None:
+        causal_mask = attention_mask[:, :, :, : key_states.shape[-2]]
+        attn_weights = attn_weights + causal_mask
+    attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query.dtype)
+    attn_weights = nn.functional.dropout(attn_weights, p=dropout, training=module.training)
+    attn_output = torch.matmul(attn_weights, value_states)
+    attn_output = attn_output.transpose(1, 2).contiguous()
+    return attn_output, attn_weights
+def calc_router_weights(
+    router_logits: torch.Tensor,
+    router_bias: Optional[torch.Tensor],
+    score_func: str,
+    top_k: int,
+    scaling_factor: Optional[float],
+) -> tuple[torch.Tensor, torch.Tensor]:
+    """Return native-XLLM-compatible routing weights and selected experts.
+    XLLM applies router bias only to the values used for top-k selection. The
+    selected routes are still weighted by the original router probabilities,
+    then optionally normalized and scaled.
+    """
+    if score_func == "softmax":
+        routing_scores = F.softmax(router_logits, dim=-1, dtype=torch.float32)
+    elif score_func == "sigmoid":
+        routing_scores = torch.sigmoid(router_logits.to(torch.float32))
+    else:
+        raise ValueError(f"Unsupported router score function: {score_func}")
+    selection_scores = routing_scores
+    if router_bias is not None:
+        selection_scores = selection_scores + router_bias.to(selection_scores)
+    selected_indices = torch.topk(selection_scores, top_k, dim=-1).indices
+    routing_weights = torch.gather(routing_scores, dim=-1, index=selected_indices)
+    if top_k > 1:
+        routing_weights = routing_weights / routing_weights.sum(dim=-1, keepdim=True)
+    if scaling_factor is not None:
+        routing_weights = routing_weights * scaling_factor
+    return routing_weights, selected_indices
+def combine_routed_experts(
+    hidden_states: torch.Tensor,
+    routing_weights: torch.Tensor,
+    selected_indices: torch.Tensor,
+    experts: nn.ModuleList,
+    activation: Optional[Callable[[torch.Tensor], torch.Tensor]] = None,
+) -> torch.Tensor:
+    num_tokens, hidden_dim = hidden_states.shape
+    final_hidden_states = torch.zeros(
+        (num_tokens, experts[0].out_features),
+        dtype=hidden_states.dtype,
+        device=hidden_states.device,
+    )
+    expert_mask = torch.nn.functional.one_hot(
+        selected_indices, num_classes=len(experts)
+    ).permute(2, 1, 0)
+    for expert_idx in torch.nonzero(expert_mask.sum(dim=(-1, -2)), as_tuple=False).flatten():
+        topk_positions, token_positions = torch.where(expert_mask[int(expert_idx)])
+        expert_states = experts[int(expert_idx)](hidden_states[token_positions])
+        if activation is not None:
+            expert_states = activation(expert_states)
+        expert_states = expert_states * routing_weights[token_positions, topk_positions, None].to(expert_states.dtype)
+        final_hidden_states.index_add_(0, token_positions, expert_states.to(hidden_states.dtype))
+    return final_hidden_states
+class K2HorizonAttention(nn.Module):
+    """Multi-headed attention from 'Attention Is All You Need' paper"""
+    def __init__(self, config: K2HorizonConfig, layer_idx: int):
+        super().__init__()
+        self.config = config
+        self.layer_idx = layer_idx
+        self.head_dim = getattr(config, "head_dim", config.hidden_size // config.num_attention_heads)
+        self.num_key_value_groups = config.num_attention_heads // config.num_key_value_heads
+        self.scaling = self.head_dim ** -0.5
+        self.attention_dropout = config.attention_dropout
+        self.is_causal = True
+        self.rope_head_dim = self.head_dim if config.rope_head_dim is None else config.rope_head_dim
+        self.q_proj = nn.Linear(
+            config.hidden_size, config.num_attention_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.k_proj = nn.Linear(
+            config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.v_proj = nn.Linear(
+            config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.o_proj = nn.Linear(
+            config.num_attention_heads * self.head_dim, config.hidden_size, bias=config.attention_bias
+        )
+        self.gate_func = config.attention_gate_func
+        if self.gate_func is not None:
+            self.gate_proj = nn.Linear(
+                config.hidden_size,
+                config.num_attention_heads * self.head_dim,
+                bias=False)
+        if config.query_key_norm:
+            self.q_norm = K2HorizonRMSNorm(
+                hidden_size=config.num_attention_heads * self.head_dim,
+                n_groups=config.num_attention_heads,
+                eps=config.rms_norm_eps)
+            self.k_norm = K2HorizonRMSNorm(
+                hidden_size=config.num_key_value_heads * self.head_dim,
+                n_groups=config.num_key_value_heads,
+                eps=config.rms_norm_eps)
+        self.sliding_window = getattr(config, "sliding_window", None)
+    @deprecate_kwarg("past_key_value", new_name="past_key_values", version="4.58")
+    def forward(
+            self,
+            hidden_states: torch.Tensor,
+            position_embeddings: tuple[torch.Tensor, torch.Tensor],
+            attention_mask: Optional[torch.Tensor],
+            past_key_values: Optional[Cache] = None,
+            cache_position: Optional[torch.LongTensor] = None,
+            **kwargs: Unpack[FlashAttentionKwargs],
+    ) -> tuple[torch.Tensor, Optional[torch.Tensor]]:
+        input_shape = hidden_states.shape[:-1]
+        hidden_shape = (*input_shape, -1, self.head_dim)
+        if self.config.query_key_norm:
+            query_states = self.q_norm(self.q_proj(hidden_states)).view(hidden_shape).transpose(1, 2)
+            key_states = self.k_norm(self.k_proj(hidden_states)).view(hidden_shape).transpose(1, 2)
+        else:
+            query_states = self.q_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+            key_states = self.k_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+        value_states = self.v_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+        cos, sin = position_embeddings
+        if self.rope_head_dim == self.head_dim:
+            query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin)
+        else:
+            query_states, query_states_ = torch.split(
+                split_to_interleaved(query_states),
+                split_size_or_sections=[self.rope_head_dim, self.head_dim - self.rope_head_dim],
+                dim=-1)
+            key_states, key_states_ = torch.split(
+                split_to_interleaved(key_states),
+                split_size_or_sections=[self.rope_head_dim, self.head_dim - self.rope_head_dim],
+                dim=-1)
+            query_states, key_states = apply_rotary_pos_emb(
+                interleaved_to_split(query_states),
+                interleaved_to_split(key_states),
+                cos,
+                sin)
+            query_states = interleaved_to_split(torch.cat(
+                [split_to_interleaved(query_states), query_states_], dim=-1))
+            key_states = interleaved_to_split(torch.cat(
+                [split_to_interleaved(key_states), key_states_], dim=-1))
+        if past_key_values is not None:
+            # sin and cos are specific to RoPE models; cache_position needed for the static cache
+            cache_kwargs = {"sin": sin, "cos": cos, "cache_position": cache_position}
+            key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx, cache_kwargs)
+        attention_interface: Callable = eager_attention_forward
+        if self.config._attn_implementation != "eager":
+            attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation]
+        attn_output, attn_weights = attention_interface(
+            self,
+            query_states,
+            key_states,
+            value_states,
+            attention_mask,
+            dropout=0.0 if not self.training else self.attention_dropout,
+            scaling=self.scaling,
+            sliding_window=self.sliding_window,  # diff with Llama
+            **kwargs,
+        )
+        if self.gate_func is not None:
+            gate = self.gate_proj(hidden_states).view(
+                input_shape + (-1, self.head_dim))
+            if self.gate_func == 'silu':
+                gate = F.silu(gate)
+            else:
+                assert self.gate_func == 'softplus'
+                gate = F.softplus(gate, beta=math.log(2))
+            attn_output = attn_output * gate
+        attn_output = attn_output.reshape(*input_shape, -1).contiguous()
+        attn_output = self.o_proj(attn_output)
+        return attn_output, attn_weights
+def apply_rotary_pos_emb_xllm(q, k, freqs_cis):
+    if q.shape[-1] % 2 != 0 or k.shape[-1] % 2 != 0:
+        raise ValueError(f"RoPE dimensions must be even, got q={q.shape[-1]} and k={k.shape[-1]}")
+    q_ = torch.view_as_complex(q.float().reshape(*q.shape[:-1], -1, 2))
+    k_ = torch.view_as_complex(k.float().reshape(*k.shape[:-1], -1, 2))
+    if freqs_cis.ndim == 2:
+        freqs_cis = freqs_cis.unsqueeze(1)
+    elif freqs_cis.ndim == 3:
+        freqs_cis = freqs_cis.unsqueeze(2)
+    else:
+        raise ValueError(f"Unsupported freqs_cis shape: {tuple(freqs_cis.shape)}")
+    if freqs_cis.shape[-1] != q_.shape[-1] or freqs_cis.shape[-1] != k_.shape[-1]:
+        raise ValueError(
+            "RoPE frequency dimension mismatch: "
+            f"q_rope_dim={q.shape[-1]}, k_rope_dim={k.shape[-1]}, "
+            f"q_complex_dim={q_.shape[-1]}, k_complex_dim={k_.shape[-1]}, "
+            f"freqs_complex_dim={freqs_cis.shape[-1]}, freqs_shape={tuple(freqs_cis.shape)}"
+        )
+    q_embed = torch.view_as_real(q_ * freqs_cis).flatten(-2).to(q.dtype)
+    k_embed = torch.view_as_real(k_ * freqs_cis).flatten(-2).to(k.dtype)
+    return q_embed, k_embed
+class K2HorizonMoVAAttention(nn.Module):
+    """MoVA attention with routed value experts and optional post-attention gate."""
+    def __init__(self, config: K2HorizonConfig, layer_idx: int):
+        super().__init__()
+        self.config = config
+        self.layer_idx = layer_idx
+        self.head_dim = getattr(config, "head_dim", config.hidden_size // config.num_attention_heads)
+        self.num_key_value_groups = config.num_attention_heads // config.num_key_value_heads
+        self.scaling = self.head_dim**-0.5
+        self.attention_dropout = config.attention_dropout
+        self.is_causal = True
+        self.num_experts_per_tok = config.mova_num_experts_per_tok
+        self.router_score_func = config.router_score_func
+        self.router_scaling_factor = config.router_scaling_factor
+        self.gate_func = config.attention_gate_func
+        self.rope_head_dim = self.head_dim if config.rope_head_dim is None else config.rope_head_dim
+        self.q_proj = nn.Linear(
+            config.hidden_size, config.num_attention_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.k_proj = nn.Linear(
+            config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.o_proj = nn.Linear(
+            config.num_attention_heads * self.head_dim, config.hidden_size, bias=config.attention_bias
+        )
+        self.v_router = nn.Linear(
+            config.hidden_size,
+            config.mova_num_experts,
+            bias=config.moe_gate_bias)
+        self.v_experts = nn.ModuleList([
+            nn.Linear(
+                config.hidden_size,
+                config.num_key_value_heads * self.head_dim,
+                bias=False
+            ) for _ in range(config.mova_num_experts)
+        ])
+        if self.gate_func is not None:
+            self.gate_proj = nn.Linear(
+                config.hidden_size,
+                config.num_attention_heads * self.head_dim,
+                bias=False)
+        if config.query_key_norm:
+            self.q_norm = K2HorizonRMSNorm(
+                hidden_size=config.num_attention_heads * self.head_dim,
+                n_groups=config.num_attention_heads,
+                eps=config.rms_norm_eps,
+            )
+            self.k_norm = K2HorizonRMSNorm(
+                hidden_size=config.num_key_value_heads * self.head_dim,
+                n_groups=config.num_key_value_heads,
+                eps=config.rms_norm_eps,
+            )
+        self.sliding_window = getattr(config, "sliding_window", None)
+    @deprecate_kwarg("past_key_value", new_name="past_key_values", version="4.58")
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        position_embeddings: torch.Tensor,
+        attention_mask: Optional[torch.Tensor],
+        past_key_values: Optional[Cache] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        **kwargs: Unpack[FlashAttentionKwargs],
+    ) -> tuple[torch.Tensor, Optional[torch.Tensor]]:
+        input_shape = hidden_states.shape[:-1]
+        hidden_shape = (*input_shape, -1, self.head_dim)
+        flat_hidden_states = hidden_states.reshape(-1, hidden_states.shape[-1])
+        # Match native MOVAttention router semantics exactly: compute logits with
+        # the weight-only linear and apply router bias only to selection scores.
+        router_logits = F.linear(flat_hidden_states, self.v_router.weight)
+        routing_weights, selected_values = calc_router_weights(
+            router_logits=router_logits,
+            router_bias=self.v_router.bias,
+            score_func=self.router_score_func,
+            top_k=self.num_experts_per_tok,
+            scaling_factor=self.router_scaling_factor,
+        )
+        mixed_value_states = combine_routed_experts(
+            hidden_states=flat_hidden_states,
+            routing_weights=routing_weights,
+            selected_indices=selected_values,
+            experts=self.v_experts,
+            activation=F.silu)
+        value_states = mixed_value_states.view(hidden_shape).transpose(1, 2)
+        if self.config.query_key_norm:
+            query_states = self.q_norm(self.q_proj(hidden_states)).view(hidden_shape).transpose(1, 2)
+            key_states = self.k_norm(self.k_proj(hidden_states)).view(hidden_shape).transpose(1, 2)
+        else:
+            query_states = self.q_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+            key_states = self.k_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+        cos, sin = position_embeddings
+        if self.rope_head_dim == self.head_dim:
+            query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin)
+        else:
+            query_states, query_states_ = torch.split(
+                split_to_interleaved(query_states),
+                split_size_or_sections=[self.rope_head_dim, self.head_dim - self.rope_head_dim],
+                dim=-1)
+            key_states, key_states_ = torch.split(
+                split_to_interleaved(key_states),
+                split_size_or_sections=[self.rope_head_dim, self.head_dim - self.rope_head_dim],
+                dim=-1)
+            query_states, key_states = apply_rotary_pos_emb(
+                interleaved_to_split(query_states),
+                interleaved_to_split(key_states),
+                cos,
+                sin)
+            query_states = interleaved_to_split(torch.cat(
+                [split_to_interleaved(query_states), query_states_], dim=-1))
+            key_states = interleaved_to_split(torch.cat(
+                [split_to_interleaved(key_states), key_states_], dim=-1))
+        if past_key_values is not None:
+            cache_kwargs = {"cache_position": cache_position}
+            key_states, value_states = past_key_values.update(
+                key_states, value_states, self.layer_idx, cache_kwargs
+            )
+        attention_interface: Callable = eager_attention_forward
+        if self.config._attn_implementation != "eager":
+            attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation]
+        attn_output, attn_weights = attention_interface(
+            self,
+            query_states,
+            key_states,
+            value_states,
+            attention_mask,
+            dropout=0.0 if not self.training else self.attention_dropout,
+            scaling=self.scaling,
+            sliding_window=self.sliding_window,
+            **kwargs,
+        )
+        if self.gate_func is not None:
+            gate = self.gate_proj(hidden_states).view(input_shape + (-1, self.head_dim))
+            if self.gate_func == 'silu':
+                gate = F.silu(gate)
+            else:
+                assert self.gate_func == 'softplus'
+                gate = F.softplus(gate, beta=math.log(2))
+            attn_output = attn_output * gate
+        attn_output = attn_output.reshape(*input_shape, -1).contiguous()
+        attn_output = self.o_proj(attn_output)
+        return attn_output, attn_weights
+class K2HorizonMLP(nn.Module):
+    def __init__(self, config, intermediate_size=None):
+        super().__init__()
+        self.config = config
+        self.hidden_size = config.hidden_size
+        self.intermediate_size = intermediate_size if intermediate_size is not None else config.intermediate_size
+        self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
+        self.act_fn = ACT2FN[config.hidden_act]
+    def forward(self, x):
+        down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
+        return down_proj
+class K2HorizonSparseMoeBlock(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.num_experts = config.num_experts
+        self.top_k = config.num_experts_per_tok
+        self.norm_topk_prob = config.norm_topk_prob
+        self.num_shared_experts = config.num_shared_experts
+        self.router_score_func = config.router_score_func
+        self.router_scaling_factor = config.router_scaling_factor
+        # gating
+        self.gate = nn.Linear(config.hidden_size, config.num_experts, bias=config.moe_gate_bias)
+        self.experts = nn.ModuleList(
+            [K2HorizonMLP(config, intermediate_size=config.moe_intermediate_size) for _ in range(self.num_experts)]
+        )
+        if config.num_shared_experts > 0:
+            self.shared_experts = K2HorizonMLP(
+                config=config,
+                intermediate_size=config.moe_intermediate_size * config.num_shared_experts)
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        """ """
+        residuals = hidden_states
+        batch_size, sequence_length, hidden_dim = hidden_states.shape
+        hidden_states = hidden_states.view(-1, hidden_dim)
+        # router_logits: (batch * sequence_length, n_experts)
+        # router_logits = self.gate(hidden_states)
+        router_logits = F.linear(hidden_states, self.gate.weight)
+        if self.router_score_func == "softmax":
+            routing_weights = F.softmax(router_logits, dim=1, dtype=torch.float)
+        else:
+            assert self.router_score_func == "sigmoid"
+            routing_weights = F.sigmoid(router_logits.to(torch.float32))
+        routing_weights_for_choice = routing_weights
+        if self.gate.bias is not None:
+            routing_weights_for_choice = routing_weights + self.gate.bias.to(routing_weights.dtype)
+        _, selected_experts = torch.topk(routing_weights_for_choice, self.top_k, dim=-1)
+        routing_weights = torch.gather(routing_weights, dim=-1, index=selected_experts)
+        if self.norm_topk_prob:  # only diff with mixtral sparse moe block!
+            routing_weights /= routing_weights.sum(dim=-1, keepdim=True)
+        routing_weights = routing_weights * self.router_scaling_factor
+        # we cast back to the input dtype
+        routing_weights = routing_weights.to(hidden_states.dtype)
+        final_hidden_states = torch.zeros(
+            (batch_size * sequence_length, hidden_dim), dtype=hidden_states.dtype, device=hidden_states.device
+        )
+        # One hot encode the selected experts to create an expert mask
+        # this will be used to easily index which expert is going to be sollicitated
+        expert_mask = torch.nn.functional.one_hot(selected_experts, num_classes=self.num_experts).permute(2, 1, 0)
+        # Loop over all available experts in the model and perform the computation on each expert
+        expert_hit = torch.greater(expert_mask.sum(dim=(-1, -2)), 0).nonzero()
+        for expert_idx in expert_hit:
+            expert_layer = self.experts[expert_idx]
+            idx, top_x = torch.where(expert_mask[expert_idx].squeeze(0))
+            # Index the correct hidden states and compute the expert hidden state for
+            # the current expert. We need to make sure to multiply the output hidden
+            # states by `routing_weights` on the corresponding tokens (top-1 and top-2)
+            current_state = hidden_states[None, top_x].reshape(-1, hidden_dim)
+            current_hidden_states = expert_layer(current_state) * routing_weights[top_x, idx, None]
+            # However `index_add_` only support torch tensors for indexing so we'll use
+            # the `top_x` tensor here.
+            final_hidden_states.index_add_(0, top_x, current_hidden_states.to(hidden_states.dtype))
+        final_hidden_states = final_hidden_states.reshape(batch_size, sequence_length, hidden_dim)
+        if self.num_shared_experts > 0:
+            final_hidden_states = final_hidden_states + self.shared_experts(residuals)
+        return final_hidden_states, router_logits
+# @use_kernel_forward_from_hub("RMSNorm")
+class K2HorizonRMSNorm(nn.Module):
+    def __init__(self, hidden_size: int, n_groups: int, eps=1e-6):
+        """
+        K2HorizonRMSNorm is equivalent to T5LayerNorm
+        """
+        super().__init__()
+        self.n_groups = n_groups
+        self.hidden_size = hidden_size
+        assert hidden_size % n_groups == 0
+        self.weight = nn.Parameter(torch.ones(hidden_size))
+        self.variance_epsilon = eps
+    def forward(self, hidden_states):
+        input_dtype = hidden_states.dtype
+        hidden_states = hidden_states.to(torch.float32)
+        hidden_states = hidden_states.reshape(*hidden_states.shape[:-1], self.n_groups, -1)
+        variance = hidden_states.pow(2).mean(-1, keepdim=True)
+        hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon)
+        hidden_states = hidden_states.reshape(*hidden_states.shape[:-2], -1)
+        hidden_states = self.weight * hidden_states
+        return hidden_states.to(input_dtype)
+    def extra_repr(self):
+        return f"{tuple(self.weight.shape)}, eps={self.variance_epsilon}"
+class K2HorizonDecoderLayer(GradientCheckpointingLayer):
+    def __init__(self, config: K2HorizonConfig, layer_idx: int):
+        super().__init__()
+        self.hidden_size = config.hidden_size
+        is_sparse_layer = (layer_idx not in config.mlp_only_layers) and (
+                config.num_experts > 0 and (layer_idx + 1) % config.decoder_sparse_step == 0)
+        if is_sparse_layer and config.mova_num_experts > 0:
+            self.self_attn = K2HorizonMoVAAttention(config=config, layer_idx=layer_idx)
+        else:
+            self.self_attn = K2HorizonAttention(config, layer_idx)
+        if is_sparse_layer:
+            self.mlp = K2HorizonSparseMoeBlock(config)
+        else:
+            self.mlp = K2HorizonMLP(config, intermediate_size=config.intermediate_size)
+        assert config.hidden_size % config.layernorm_num_groups == 0
+        self.input_layernorm = K2HorizonRMSNorm(
+            hidden_size=config.hidden_size,
+            n_groups=config.layernorm_num_groups,
+            eps=config.rms_norm_eps)
+        self.post_attention_layernorm = K2HorizonRMSNorm(
+            hidden_size=config.hidden_size,
+            n_groups=config.layernorm_num_groups,
+            eps=config.rms_norm_eps)
+    @deprecate_kwarg("past_key_value", new_name="past_key_values", version="4.58")
+    def forward(
+            self,
+            hidden_states: torch.Tensor,
+            position_embeddings: tuple[torch.Tensor, torch.Tensor],
+            attention_mask: Optional[torch.Tensor] = None,
+            position_ids: Optional[torch.LongTensor] = None,
+            past_key_values: Optional[Cache] = None,
+            cache_position: Optional[torch.LongTensor] = None,
+            **kwargs: Unpack[FlashAttentionKwargs],
+    ) -> torch.FloatTensor:
+        """
+        Args:
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
+            attention_mask (`torch.FloatTensor`, *optional*): attention mask of size
+                `(batch, sequence_length)` where padding elements are indicated by 0.
+            output_attentions (`bool`, *optional*):
+                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
+                returned tensors for more detail.
+            output_router_logits (`bool`, *optional*):
+                Whether or not to return the logits of all the routers. They are useful for computing the router loss,
+                and should not be returned during inference.
+            use_cache (`bool`, *optional*):
+                If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
+                (see `past_key_values`).
+            past_key_values (`Cache`, *optional*): cached past key and value projection states
+            cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*):
+                Indices depicting the position of the input sequence tokens in the sequence.
+            position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*):
+                Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`,
+                with `head_dim` being the embedding dimension of each attention head.
+            kwargs (`dict`, *optional*):
+                Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code
+                into the model
+        """
+        residual = hidden_states
+        hidden_states = self.input_layernorm(hidden_states)
+        # Self Attention
+        hidden_states, _ = self.self_attn(
+            hidden_states=hidden_states,
+            position_embeddings=position_embeddings,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            cache_position=cache_position,
+            **kwargs,
+        )
+        hidden_states = residual + hidden_states
+        # Fully Connected
+        residual = hidden_states
+        hidden_states = self.post_attention_layernorm(hidden_states)
+        hidden_states = self.mlp(hidden_states)
+        # For the MoE layers, we need to unpack
+        if isinstance(hidden_states, tuple):
+            hidden_states, _ = hidden_states
+        hidden_states = residual + hidden_states
+        return hidden_states
+class K2HorizonRotaryEmbedding(nn.Module):
+    inv_freq: torch.Tensor  # fix linting for `register_buffer`
+    def __init__(self, config: K2HorizonConfig, device=None):
+        super().__init__()
+        self.max_seq_len_cached = config.max_position_embeddings
+        self.original_max_seq_len = config.max_position_embeddings
+        self.config = config
+        self.rope_type = self.config.rope_parameters["rope_type"]
+        rope_init_fn: Callable = self.compute_default_rope_parameters
+        if self.rope_type != "default":
+            rope_init_fn = ROPE_INIT_FUNCTIONS[self.rope_type]
+        inv_freq, self.attention_scaling = rope_init_fn(self.config, device)
+        self.register_buffer("inv_freq", inv_freq, persistent=False)
+        self.register_buffer("original_inv_freq", inv_freq.clone(), persistent=False)
+    @staticmethod
+    def compute_default_rope_parameters(
+        config: K2HorizonConfig | None = None,
+        device: Optional["torch.device"] = None,
+        seq_len: int | None = None,
+    ) -> tuple["torch.Tensor", float]:
+        """
+        Computes the inverse frequencies according to the original RoPE implementation
+        Args:
+            config ([`~transformers.PreTrainedConfig`]):
+                The model configuration.
+            device (`torch.device`):
+                The device to use for initialization of the inverse frequencies.
+            seq_len (`int`, *optional*):
+                The current sequence length. Unused for this type of RoPE.
+        Returns:
+            Tuple of (`torch.Tensor`, `float`), containing the inverse frequencies for the RoPE embeddings and the
+            post-processing scaling factor applied to the computed cos/sin (unused in this type of RoPE).
+        """
+        base = config.rope_parameters["rope_theta"]
+        # dim = getattr(config, "head_dim", None) or config.hidden_size // config.num_attention_heads
+        dim = (
+            config.rope_head_dim
+            if config.rope_head_dim is not None
+            else getattr(config, "head_dim", None) or config.hidden_size // config.num_attention_heads
+        )
+        attention_factor = 1.0  # Unused in this type of RoPE
+        # Compute the inverse frequencies
+        inv_freq = 1.0 / (
+            base ** (torch.arange(0, dim, 2, dtype=torch.int64).to(device=device, dtype=torch.float) / dim)
+        )
+        return inv_freq, attention_factor
+    @torch.no_grad()
+    @dynamic_rope_update  # power user: used with advanced RoPE types (e.g. dynamic rope)
+    def forward(self, x, position_ids):
+        inv_freq_expanded = self.inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1).to(x.device)
+        position_ids_expanded = position_ids[:, None, :].float()
+        device_type = x.device.type if isinstance(x.device.type, str) and x.device.type != "mps" else "cpu"
+        with maybe_autocast(device_type=device_type, enabled=False):  # Force float32
+            freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2)
+            emb = torch.cat((freqs, freqs), dim=-1)
+            cos = emb.cos() * self.attention_scaling
+            sin = emb.sin() * self.attention_scaling
+        return cos.to(dtype=x.dtype), sin.to(dtype=x.dtype)
+@auto_docstring
+class K2HorizonPreTrainedModel(PreTrainedModel):
+    config: K2HorizonConfig
+    base_model_prefix = "model"
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["K2HorizonDecoderLayer"]
+    _skip_keys_device_placement = ["past_key_values"]
+    _supports_flash_attn = True
+    _supports_sdpa = True
+    _supports_flex_attn = True
+    _can_compile_fullgraph = False  # MoE models don't work with torch.compile (`torch.where(condition)` not supported)
+    _supports_attention_backend = True
+    _can_record_outputs = {
+        "router_logits": OutputRecorder(K2HorizonSparseMoeBlock, index=1),
+        "hidden_states": K2HorizonDecoderLayer,
+        "attentions": K2HorizonAttention,
+    }
+@auto_docstring
+class K2HorizonModel(K2HorizonPreTrainedModel):
+    def __init__(self, config: K2HorizonConfig):
+        super().__init__(config)
+        # self.padding_idx = config.pad_token_id
+        self.padding_idx = getattr(config, "padding_idx", None)
+        self.vocab_size = config.vocab_size
+        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
+        self.layers = nn.ModuleList(
+            [K2HorizonDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]
+        )
+        assert config.hidden_size % config.layernorm_num_groups == 0
+        self.norm = K2HorizonRMSNorm(
+            hidden_size=config.hidden_size,
+            n_groups=config.layernorm_num_groups,
+            eps=config.rms_norm_eps)
+        self.rotary_emb = K2HorizonRotaryEmbedding(config=config)
+        self.gradient_checkpointing = False
+        # Initialize weights and apply final processing
+        self.post_init()
+    @auto_docstring
+    def forward(
+            self,
+            input_ids: Optional[torch.LongTensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            position_ids: Optional[torch.LongTensor] = None,
+            past_key_values: Optional[Cache] = None,
+            inputs_embeds: Optional[torch.FloatTensor] = None,
+            use_cache: Optional[bool] = None,
+            cache_position: Optional[torch.LongTensor] = None,
+            **kwargs: Unpack[TransformersKwargs],
+    ) -> MoeModelOutputWithPast:
+        if (input_ids is None) ^ (inputs_embeds is not None):
+            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
+        if use_cache and past_key_values is None:
+            past_key_values = DynamicCache(config=self.config)
+        if inputs_embeds is None:
+            inputs_embeds = self.embed_tokens(input_ids)
+        if cache_position is None:
+            past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
+            cache_position = torch.arange(
+                past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device
+            )
+        if position_ids is None:
+            position_ids = cache_position.unsqueeze(0)
+        mask_function = create_causal_mask if self.config.sliding_window is None else create_sliding_window_causal_mask
+        causal_mask = mask_function(
+            config=self.config,
+            inputs_embeds=inputs_embeds,
+            attention_mask=attention_mask,
+            past_key_values=past_key_values,
+            position_ids=position_ids,
+        )
+        hidden_states = inputs_embeds
+        # create position embeddings to be shared across the decoder layers
+        position_embeddings = self.rotary_emb(hidden_states, position_ids)
+        for layer_idx, decoder_layer in enumerate(self.layers[: self.config.num_hidden_layers]):
+            hidden_states = decoder_layer(
+                hidden_states,
+                position_embeddings=position_embeddings,
+                attention_mask=causal_mask,
+                position_ids=position_ids,
+                past_key_values=past_key_values,
+                use_cache=use_cache,
+                cache_position=cache_position,
+                **kwargs,
+            )
+        hidden_states = self.norm(hidden_states)
+        return MoeModelOutputWithPast(  # only diff with Mistral is the output type, we need MoE
+            last_hidden_state=hidden_states,
+            past_key_values=past_key_values,
+        )
+def load_balancing_loss_func(
+        gate_logits: Union[torch.Tensor, tuple[torch.Tensor], None],
+        num_experts: Optional[int] = None,
+        top_k=2,
+        attention_mask: Optional[torch.Tensor] = None,
+) -> Union[torch.Tensor, int]:
+    r"""
+    Computes auxiliary load balancing loss as in Switch Transformer - implemented in Pytorch.
+    See Switch Transformer (https://huggingface.co/papers/2101.03961) for more details. This function implements the loss
+    function presented in equations (4) - (6) of the paper. It aims at penalizing cases where the routing between
+    experts is too unbalanced.
+    Args:
+        gate_logits:
+            Logits from the `gate`, should be a tuple of model.config.num_hidden_layers tensors of
+            shape [batch_size X sequence_length, num_experts].
+        num_experts:
+            Number of experts
+        top_k:
+            The number of experts to route per-token, can be also interpreted as the `top-k` routing
+            parameter.
+        attention_mask (`torch.Tensor`, *optional*):
+            The attention_mask used in forward function
+            shape [batch_size X sequence_length] if not None.
+    Returns:
+        The auxiliary loss.
+    """
+    if gate_logits is None or not isinstance(gate_logits, tuple):
+        return 0
+    if isinstance(gate_logits, tuple):
+        compute_device = gate_logits[0].device
+        concatenated_gate_logits = torch.cat([layer_gate.to(compute_device) for layer_gate in gate_logits], dim=0)
+    routing_weights = torch.nn.functional.softmax(concatenated_gate_logits, dim=-1)
+    _, selected_experts = torch.topk(routing_weights, top_k, dim=-1)
+    expert_mask = torch.nn.functional.one_hot(selected_experts, num_experts)
+    if attention_mask is None:
+        # Compute the percentage of tokens routed to each experts
+        tokens_per_expert = torch.mean(expert_mask.float(), dim=0)
+        # Compute the average probability of routing to these experts
+        router_prob_per_expert = torch.mean(routing_weights, dim=0)
+    else:
+        batch_size, sequence_length = attention_mask.shape
+        num_hidden_layers = concatenated_gate_logits.shape[0] // (batch_size * sequence_length)
+        # Compute the mask that masks all padding tokens as 0 with the same shape of expert_mask
+        expert_attention_mask = (
+            attention_mask[None, :, :, None, None]
+            .expand((num_hidden_layers, batch_size, sequence_length, top_k, num_experts))
+            .reshape(-1, top_k, num_experts)
+            .to(compute_device)
+        )
+        # Compute the percentage of tokens routed to each experts
+        tokens_per_expert = torch.sum(expert_mask.float() * expert_attention_mask, dim=0) / torch.sum(
+            expert_attention_mask, dim=0
+        )
+        # Compute the mask that masks all padding tokens as 0 with the same shape of tokens_per_expert
+        router_per_expert_attention_mask = (
+            attention_mask[None, :, :, None]
+            .expand((num_hidden_layers, batch_size, sequence_length, num_experts))
+            .reshape(-1, num_experts)
+            .to(compute_device)
+        )
+        # Compute the average probability of routing to these experts
+        router_prob_per_expert = torch.sum(routing_weights * router_per_expert_attention_mask, dim=0) / torch.sum(
+            router_per_expert_attention_mask, dim=0
+        )
+    overall_loss = torch.sum(tokens_per_expert * router_prob_per_expert.unsqueeze(0))
+    return overall_loss * num_experts
+@auto_docstring
+class K2HorizonForCausalLM(K2HorizonPreTrainedModel, GenerationMixin):
+    # _tied_weights_keys = ["lm_head.weight"]
+    # _tp_plan = {"lm_head": "colwise_rep"}
+    # _pp_plan = {"lm_head": (["hidden_states"], ["logits"])}
+    def __init__(self, config):
+        super().__init__(config)
+        self.model = K2HorizonModel(config)
+        self.vocab_size = config.vocab_size
+        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+        self.router_aux_loss_coef = config.router_aux_loss_coef
+        self.num_experts = config.num_experts
+        self.num_experts_per_tok = config.num_experts_per_tok
+        # Initialize weights and apply final processing
+        self.post_init()
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+            self,
+            input_ids: Optional[torch.LongTensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            position_ids: Optional[torch.LongTensor] = None,
+            past_key_values: Optional[Cache] = None,
+            inputs_embeds: Optional[torch.FloatTensor] = None,
+            labels: Optional[torch.LongTensor] = None,
+            use_cache: Optional[bool] = None,
+            output_router_logits: Optional[bool] = None,
+            cache_position: Optional[torch.LongTensor] = None,
+            logits_to_keep: Union[int, torch.Tensor] = 0,
+            **kwargs: Unpack[TransformersKwargs],
+    ) -> MoeCausalLMOutputWithPast:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
+            config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
+        Example:
+        ```python
+        >>> from transformers import AutoTokenizer, K2HorizonForCausalLM
+        >>> model = K2HorizonForCausalLM.from_pretrained("Qwen/Qwen3-MoE-15B-A2B")
+        >>> tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-MoE-15B-A2B")
+        >>> prompt = "Hey, are you conscious? Can you talk to me?"
+        >>> inputs = tokenizer(prompt, return_tensors="pt")
+        >>> # Generate
+        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
+        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
+        ```"""
+        output_router_logits = (
+            output_router_logits if output_router_logits is not None else self.config.output_router_logits
+        )
+        # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
+        outputs: MoeModelOutputWithPast = self.model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_router_logits=output_router_logits,
+            cache_position=cache_position,
+            **kwargs,
+        )
+        hidden_states = outputs.last_hidden_state
+        # Only compute necessary logits, and do not upcast them to float if we are not computing the loss
+        slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep
+        logits = self.lm_head(hidden_states[:, slice_indices, :])
+        loss = None
+        if labels is not None:
+            loss = self.loss_function(logits, labels, self.vocab_size, **kwargs)
+        aux_loss = None
+        if output_router_logits:
+            aux_loss = load_balancing_loss_func(
+                outputs.router_logits,
+                self.num_experts,
+                self.num_experts_per_tok,
+                attention_mask,
+            )
+            if labels is not None:
+                loss += self.router_aux_loss_coef * aux_loss.to(loss.device)  # make sure to reside in the same device
+        return MoeCausalLMOutputWithPast(
+            loss=loss,
+            aux_loss=aux_loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+            router_logits=outputs.router_logits,
+        )
+class K2HorizonForSequenceClassification(GenericForSequenceClassification, K2HorizonPreTrainedModel):
+    pass
+class K2HorizonForTokenClassification(GenericForTokenClassification, K2HorizonPreTrainedModel):
+    pass
+class K2HorizonForQuestionAnswering(GenericForQuestionAnswering, K2HorizonPreTrainedModel):
+    base_model_prefix = "transformer"  # For BC, where `transformer` was used instead of `model`
+__all__ = [
+    "K2HorizonForCausalLM",
+    "K2HorizonForQuestionAnswering",
+    "K2HorizonModel",
+    "K2HorizonPreTrainedModel",
+    "K2HorizonForSequenceClassification",
+    "K2HorizonForTokenClassification",
+]

oq_imatrix_report.json ADDED Viewed

	@@ -0,0 +1,364 @@

+{
+  "enabled": true,
+  "cache_path": "/Users/hermit.dave/Projects/ml/models/.oqe_imatrix/K2-Horizon-7B-Uno-merged-bf16-oQe-s128-l512.npz",
+  "cache_reused": false,
+  "entry_count": 252,
+  "calib_dataset": "oqe_code_multilingual",
+  "collection": {
+    "dataset": "oqe_code_multilingual",
+    "requested_samples": 128,
+    "seq_length": 512,
+    "adaptive": true,
+    "adaptive_step_samples": 128,
+    "adaptive_max_samples": 1024,
+    "available_samples": 1024,
+    "micro_batch_size": 32,
+    "micro_batches": 4,
+    "batch_plan": {
+      "micro_batch_size": 32,
+      "estimated_sample_bytes": 8388608,
+      "capture_budget_bytes": 306620116,
+      "system_available_bytes": 35724574720,
+      "metal_available_bytes": 62272646860,
+      "live_available_bytes": 35724574720,
+      "model_bytes": 5062563118,
+      "remaining_available_bytes": 30662011602,
+      "fits_one_sample": true,
+      "hidden_size": 4096,
+      "num_experts": 0,
+      "top_k": 0,
+      "gemma4_state_bytes": 0,
+      "num_hidden_layers": 36,
+      "num_kv_shared_layers": 0,
+      "per_layer_input_size": 0
+    },
+    "processed_samples": 128,
+    "model_class": "k2_horizon.Model",
+    "layer_model_class": "k2_horizon.K2HorizonModel",
+    "layer_model_type": "k2_horizon",
+    "layer_count": 36,
+    "layer_state_kind": "generic",
+    "successful_layer_forwards": 144,
+    "skipped_layer_forwards": 0,
+    "mtp_head_forwards": 0,
+    "installed_modules": 253,
+    "capture_module_classes": {
+      "QuantizedLinear": 253
+    },
+    "switch_capture_modules": 0,
+    "requires_expert_counts": false,
+    "coverage_sufficient": true,
+    "collection_sufficient": true,
+    "coverage": {
+      "has_expert_counts": false,
+      "expert_modules": 0,
+      "total_experts": 0,
+      "active_experts": 0,
+      "zero_count_experts": 0,
+      "active_ratio": 1.0,
+      "min_count": 0,
+      "p05_count": 0.0,
+      "p10_count": 0.0,
+      "median_count": 0.0,
+      "max_count": 0,
+      "min_required_count": 16,
+      "required_percentile": 5
+    },
+    "rounds": [
+      {
+        "processed_samples": 128,
+        "coverage_sufficient": true,
+        "collection_sufficient": true,
+        "coverage": {
+          "has_expert_counts": false,
+          "expert_modules": 0,
+          "total_experts": 0,
+          "active_experts": 0,
+          "zero_count_experts": 0,
+          "active_ratio": 1.0,
+          "min_count": 0,
+          "p05_count": 0.0,
+          "p10_count": 0.0,
+          "median_count": 0.0,
+          "max_count": 0,
+          "min_required_count": 16,
+          "required_percentile": 5
+        }
+      }
+    ]
+  },
+  "expert_coverage": {
+    "has_expert_counts": false,
+    "expert_modules": 0,
+    "total_experts": 0,
+    "active_experts": 0,
+    "zero_count_experts": 0,
+    "active_ratio": 1.0,
+    "min_count": 0,
+    "p05_count": 0.0,
+    "p10_count": 0.0,
+    "median_count": 0.0,
+    "max_count": 0,
+    "min_required_count": 16,
+    "required_percentile": 5
+  },
+  "applied": [
+    "model.layers.0.mlp.down_proj",
+    "model.layers.0.mlp.gate_proj",
+    "model.layers.0.mlp.up_proj",
+    "model.layers.0.self_attn.k_proj",
+    "model.layers.0.self_attn.o_proj",
+    "model.layers.0.self_attn.q_proj",
+    "model.layers.0.self_attn.v_proj",
+    "model.layers.1.mlp.down_proj",
+    "model.layers.1.mlp.gate_proj",
+    "model.layers.1.mlp.up_proj",
+    "model.layers.1.self_attn.k_proj",
+    "model.layers.1.self_attn.o_proj",
+    "model.layers.1.self_attn.q_proj",
+    "model.layers.1.self_attn.v_proj",
+    "model.layers.10.mlp.down_proj",
+    "model.layers.10.mlp.gate_proj",
+    "model.layers.10.mlp.up_proj",
+    "model.layers.10.self_attn.k_proj",
+    "model.layers.10.self_attn.o_proj",
+    "model.layers.10.self_attn.q_proj",
+    "model.layers.10.self_attn.v_proj",
+    "model.layers.11.mlp.down_proj",
+    "model.layers.11.mlp.gate_proj",
+    "model.layers.11.mlp.up_proj",
+    "model.layers.11.self_attn.k_proj",
+    "model.layers.11.self_attn.o_proj",
+    "model.layers.11.self_attn.q_proj",
+    "model.layers.11.self_attn.v_proj",
+    "model.layers.12.mlp.down_proj",
+    "model.layers.12.mlp.gate_proj",
+    "model.layers.12.mlp.up_proj",
+    "model.layers.12.self_attn.k_proj",
+    "model.layers.12.self_attn.o_proj",
+    "model.layers.12.self_attn.q_proj",
+    "model.layers.12.self_attn.v_proj",
+    "model.layers.13.mlp.down_proj",
+    "model.layers.13.mlp.gate_proj",
+    "model.layers.13.mlp.up_proj",
+    "model.layers.13.self_attn.k_proj",
+    "model.layers.13.self_attn.o_proj",
+    "model.layers.13.self_attn.q_proj",
+    "model.layers.13.self_attn.v_proj",
+    "model.layers.14.mlp.down_proj",
+    "model.layers.14.mlp.gate_proj",
+    "model.layers.14.mlp.up_proj",
+    "model.layers.14.self_attn.k_proj",
+    "model.layers.14.self_attn.o_proj",
+    "model.layers.14.self_attn.q_proj",
+    "model.layers.14.self_attn.v_proj",
+    "model.layers.15.mlp.down_proj",
+    "model.layers.15.mlp.gate_proj",
+    "model.layers.15.mlp.up_proj",
+    "model.layers.15.self_attn.k_proj",
+    "model.layers.15.self_attn.o_proj",
+    "model.layers.15.self_attn.q_proj",
+    "model.layers.15.self_attn.v_proj",
+    "model.layers.16.mlp.down_proj",
+    "model.layers.16.mlp.gate_proj",
+    "model.layers.16.mlp.up_proj",
+    "model.layers.16.self_attn.k_proj",
+    "model.layers.16.self_attn.o_proj",
+    "model.layers.16.self_attn.q_proj",
+    "model.layers.16.self_attn.v_proj",
+    "model.layers.17.mlp.down_proj",
+    "model.layers.17.mlp.gate_proj",
+    "model.layers.17.mlp.up_proj",
+    "model.layers.17.self_attn.k_proj",
+    "model.layers.17.self_attn.o_proj",
+    "model.layers.17.self_attn.q_proj",
+    "model.layers.17.self_attn.v_proj",
+    "model.layers.18.mlp.down_proj",
+    "model.layers.18.mlp.gate_proj",
+    "model.layers.18.mlp.up_proj",
+    "model.layers.18.self_attn.k_proj",
+    "model.layers.18.self_attn.o_proj",
+    "model.layers.18.self_attn.q_proj",
+    "model.layers.18.self_attn.v_proj",
+    "model.layers.19.mlp.down_proj",
+    "model.layers.19.mlp.gate_proj",
+    "model.layers.19.mlp.up_proj",
+    "model.layers.19.self_attn.k_proj",
+    "model.layers.19.self_attn.o_proj",
+    "model.layers.19.self_attn.q_proj",
+    "model.layers.19.self_attn.v_proj",
+    "model.layers.2.mlp.down_proj",
+    "model.layers.2.mlp.gate_proj",
+    "model.layers.2.mlp.up_proj",
+    "model.layers.2.self_attn.k_proj",
+    "model.layers.2.self_attn.o_proj",
+    "model.layers.2.self_attn.q_proj",
+    "model.layers.2.self_attn.v_proj",
+    "model.layers.20.mlp.down_proj",
+    "model.layers.20.mlp.gate_proj",
+    "model.layers.20.mlp.up_proj",
+    "model.layers.20.self_attn.k_proj",
+    "model.layers.20.self_attn.o_proj",
+    "model.layers.20.self_attn.q_proj",
+    "model.layers.20.self_attn.v_proj",
+    "model.layers.21.mlp.down_proj",
+    "model.layers.21.mlp.gate_proj",
+    "model.layers.21.mlp.up_proj",
+    "model.layers.21.self_attn.k_proj",
+    "model.layers.21.self_attn.o_proj",
+    "model.layers.21.self_attn.q_proj",
+    "model.layers.21.self_attn.v_proj",
+    "model.layers.22.mlp.down_proj",
+    "model.layers.22.mlp.gate_proj",
+    "model.layers.22.mlp.up_proj",
+    "model.layers.22.self_attn.k_proj",
+    "model.layers.22.self_attn.o_proj",
+    "model.layers.22.self_attn.q_proj",
+    "model.layers.22.self_attn.v_proj",
+    "model.layers.23.mlp.down_proj",
+    "model.layers.23.mlp.gate_proj",
+    "model.layers.23.mlp.up_proj",
+    "model.layers.23.self_attn.k_proj",
+    "model.layers.23.self_attn.o_proj",
+    "model.layers.23.self_attn.q_proj",
+    "model.layers.23.self_attn.v_proj",
+    "model.layers.24.mlp.down_proj",
+    "model.layers.24.mlp.gate_proj",
+    "model.layers.24.mlp.up_proj",
+    "model.layers.24.self_attn.k_proj",
+    "model.layers.24.self_attn.o_proj",
+    "model.layers.24.self_attn.q_proj",
+    "model.layers.24.self_attn.v_proj",
+    "model.layers.25.mlp.down_proj",
+    "model.layers.25.mlp.gate_proj",
+    "model.layers.25.mlp.up_proj",
+    "model.layers.25.self_attn.k_proj",
+    "model.layers.25.self_attn.o_proj",
+    "model.layers.25.self_attn.q_proj",
+    "model.layers.25.self_attn.v_proj",
+    "model.layers.26.mlp.down_proj",
+    "model.layers.26.mlp.gate_proj",
+    "model.layers.26.mlp.up_proj",
+    "model.layers.26.self_attn.k_proj",
+    "model.layers.26.self_attn.o_proj",
+    "model.layers.26.self_attn.q_proj",
+    "model.layers.26.self_attn.v_proj",
+    "model.layers.27.mlp.down_proj",
+    "model.layers.27.mlp.gate_proj",
+    "model.layers.27.mlp.up_proj",
+    "model.layers.27.self_attn.k_proj",
+    "model.layers.27.self_attn.o_proj",
+    "model.layers.27.self_attn.q_proj",
+    "model.layers.27.self_attn.v_proj",
+    "model.layers.28.mlp.down_proj",
+    "model.layers.28.mlp.gate_proj",
+    "model.layers.28.mlp.up_proj",
+    "model.layers.28.self_attn.k_proj",
+    "model.layers.28.self_attn.o_proj",
+    "model.layers.28.self_attn.q_proj",
+    "model.layers.28.self_attn.v_proj",
+    "model.layers.29.mlp.down_proj",
+    "model.layers.29.mlp.gate_proj",
+    "model.layers.29.mlp.up_proj",
+    "model.layers.29.self_attn.k_proj",
+    "model.layers.29.self_attn.o_proj",
+    "model.layers.29.self_attn.q_proj",
+    "model.layers.29.self_attn.v_proj",
+    "model.layers.3.mlp.down_proj",
+    "model.layers.3.mlp.gate_proj",
+    "model.layers.3.mlp.up_proj",
+    "model.layers.3.self_attn.k_proj",
+    "model.layers.3.self_attn.o_proj",
+    "model.layers.3.self_attn.q_proj",
+    "model.layers.3.self_attn.v_proj",
+    "model.layers.30.mlp.down_proj",
+    "model.layers.30.mlp.gate_proj",
+    "model.layers.30.mlp.up_proj",
+    "model.layers.30.self_attn.k_proj",
+    "model.layers.30.self_attn.o_proj",
+    "model.layers.30.self_attn.q_proj",
+    "model.layers.30.self_attn.v_proj",
+    "model.layers.31.mlp.down_proj",
+    "model.layers.31.mlp.gate_proj",
+    "model.layers.31.mlp.up_proj",
+    "model.layers.31.self_attn.k_proj",
+    "model.layers.31.self_attn.o_proj",
+    "model.layers.31.self_attn.q_proj",
+    "model.layers.31.self_attn.v_proj",
+    "model.layers.32.mlp.down_proj",
+    "model.layers.32.mlp.gate_proj",
+    "model.layers.32.mlp.up_proj",
+    "model.layers.32.self_attn.k_proj",
+    "model.layers.32.self_attn.o_proj",
+    "model.layers.32.self_attn.q_proj",
+    "model.layers.32.self_attn.v_proj",
+    "model.layers.33.mlp.down_proj",
+    "model.layers.33.mlp.gate_proj",
+    "model.layers.33.mlp.up_proj",
+    "model.layers.33.self_attn.k_proj",
+    "model.layers.33.self_attn.o_proj",
+    "model.layers.33.self_attn.q_proj",
+    "model.layers.33.self_attn.v_proj",
+    "model.layers.34.mlp.down_proj",
+    "model.layers.34.mlp.gate_proj",
+    "model.layers.34.mlp.up_proj",
+    "model.layers.34.self_attn.k_proj",
+    "model.layers.34.self_attn.o_proj",
+    "model.layers.34.self_attn.q_proj",
+    "model.layers.34.self_attn.v_proj",
+    "model.layers.35.mlp.down_proj",
+    "model.layers.35.mlp.gate_proj",
+    "model.layers.35.mlp.up_proj",
+    "model.layers.35.self_attn.k_proj",
+    "model.layers.35.self_attn.o_proj",
+    "model.layers.35.self_attn.q_proj",
+    "model.layers.35.self_attn.v_proj",
+    "model.layers.4.mlp.down_proj",
+    "model.layers.4.mlp.gate_proj",
+    "model.layers.4.mlp.up_proj",
+    "model.layers.4.self_attn.k_proj",
+    "model.layers.4.self_attn.o_proj",
+    "model.layers.4.self_attn.q_proj",
+    "model.layers.4.self_attn.v_proj",
+    "model.layers.5.mlp.down_proj",
+    "model.layers.5.mlp.gate_proj",
+    "model.layers.5.mlp.up_proj",
+    "model.layers.5.self_attn.k_proj",
+    "model.layers.5.self_attn.o_proj",
+    "model.layers.5.self_attn.q_proj",
+    "model.layers.5.self_attn.v_proj",
+    "model.layers.6.mlp.down_proj",
+    "model.layers.6.mlp.gate_proj",
+    "model.layers.6.mlp.up_proj",
+    "model.layers.6.self_attn.k_proj",
+    "model.layers.6.self_attn.o_proj",
+    "model.layers.6.self_attn.q_proj",
+    "model.layers.6.self_attn.v_proj",
+    "model.layers.7.mlp.down_proj",
+    "model.layers.7.mlp.gate_proj",
+    "model.layers.7.mlp.up_proj",
+    "model.layers.7.self_attn.k_proj",
+    "model.layers.7.self_attn.o_proj",
+    "model.layers.7.self_attn.q_proj",
+    "model.layers.7.self_attn.v_proj",
+    "model.layers.8.mlp.down_proj",
+    "model.layers.8.mlp.gate_proj",
+    "model.layers.8.mlp.up_proj",
+    "model.layers.8.self_attn.k_proj",
+    "model.layers.8.self_attn.o_proj",
+    "model.layers.8.self_attn.q_proj",
+    "model.layers.8.self_attn.v_proj",
+    "model.layers.9.mlp.down_proj",
+    "model.layers.9.mlp.gate_proj",
+    "model.layers.9.mlp.up_proj",
+    "model.layers.9.self_attn.k_proj",
+    "model.layers.9.self_attn.o_proj",
+    "model.layers.9.self_attn.q_proj",
+    "model.layers.9.self_attn.v_proj"
+  ],
+  "missing": [
+    "lm_head"
+  ],
+  "mismatched": [],
+  "zero_count_experts": 0
+}

tokenizer.json ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:838d767b7c9925ff257feb20eaa4299a8e3cc35bb3d805589c373f51d2cc3cb6
+size 20637260

tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,10 @@

+{
+  "backend": "tokenizers",
+  "bos_token": "<|ifm|begin_of_text|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|ifm|endoftext|>",
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "tokenizer_class": "TokenizersBackend"
+}