Instructions to use schoggie/K2-Horizon-MoVA-36B-A4B-W4A16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use schoggie/K2-Horizon-MoVA-36B-A4B-W4A16 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="schoggie/K2-Horizon-MoVA-36B-A4B-W4A16", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("schoggie/K2-Horizon-MoVA-36B-A4B-W4A16", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use schoggie/K2-Horizon-MoVA-36B-A4B-W4A16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "schoggie/K2-Horizon-MoVA-36B-A4B-W4A16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "schoggie/K2-Horizon-MoVA-36B-A4B-W4A16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/schoggie/K2-Horizon-MoVA-36B-A4B-W4A16

SGLang

How to use schoggie/K2-Horizon-MoVA-36B-A4B-W4A16 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "schoggie/K2-Horizon-MoVA-36B-A4B-W4A16" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "schoggie/K2-Horizon-MoVA-36B-A4B-W4A16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "schoggie/K2-Horizon-MoVA-36B-A4B-W4A16" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "schoggie/K2-Horizon-MoVA-36B-A4B-W4A16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use schoggie/K2-Horizon-MoVA-36B-A4B-W4A16 with Docker Model Runner:
```
docker model run hf.co/schoggie/K2-Horizon-MoVA-36B-A4B-W4A16
```

K2-Horizon-MoVA-36B-A4B-W4A16 / config.json

schoggie

Qwen3.8-27B W4A8 (int4 group-128 weights / int8 dynamic per-token activations)

5df543e verified 6 days ago

Raw

History Blame Contribute Delete

2.33 kB

	{
	"architectures": [
	"K2HorizonForCausalLM"
	],
	"attention_bias": false,
	"attention_dropout": 0.0,
	"attention_gate_func": "softplus",
	"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": 2560,
	"initializer_range": 0.02,
	"intermediate_size": 6144,
	"layernorm_num_groups": 2,
	"max_position_embeddings": 524288,
	"mlp_only_layers": [
	0,
	1,
	2
	],
	"model_type": "k2_horizon",
	"moe_gate_bias": true,
	"moe_intermediate_size": 768,
	"mova_num_experts": 64,
	"mova_num_experts_per_tok": 4,
	"norm_topk_prob": true,
	"num_attention_heads": 32,
	"num_experts": 100,
	"num_experts_per_tok": 8,
	"num_hidden_layers": 48,
	"num_key_value_heads": 8,
	"num_shared_experts": 1,
	"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": 2.5,
	"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_config": {
	"quant_method": "compressed-tensors",
	"format": "pack-quantized",
	"quantization_status": "compressed",
	"config_groups": {
	"group_0": {
	"targets": [
	"re:.*mlp\\.experts\\.\\d+\\.(gate\|up\|down)_proj$"
	],
	"weights": {
	"num_bits": 4,
	"type": "int",
	"symmetric": true,
	"strategy": "group",
	"group_size": 128,
	"dynamic": false,
	"block_structure": null,
	"actorder": null,
	"observer": "minmax",
	"observer_kwargs": {},
	"scale_dtype": null,
	"zp_dtype": null
	},
	"input_activations": null,
	"output_activations": null
	}
	},
	"ignore": [
	"lm_head"
	],
	"kv_cache_scheme": null
	}
	}