Instructions to use beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4 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("beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4")

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 beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4 with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4"

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": "beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4 with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4 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 "beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4"

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 beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4 with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4"

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 "beezu/GLM-4.5-Iceblink-v3-106B-A12B-mlx-mxfp4" \
  --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"

beezu commited on Mar 19

Commit

cf089b8

1 Parent(s): d418427

Upload folder using huggingface_hub

Browse files

Files changed (14) hide show

README.md +6 -2
config.json +4 -11
model-00001-of-00011.safetensors +3 -0
model-00002-of-00011.safetensors +3 -0
model-00003-of-00011.safetensors +3 -0
model-00004-of-00011.safetensors +3 -0
model-00005-of-00011.safetensors +3 -0
model-00006-of-00011.safetensors +3 -0
model-00007-of-00011.safetensors +3 -0
model-00008-of-00011.safetensors +3 -0
model-00009-of-00011.safetensors +3 -0
model-00010-of-00011.safetensors +3 -0
model-00011-of-00011.safetensors +3 -0
model.safetensors.index.json +0 -0

README.md CHANGED Viewed

@@ -20,7 +20,11 @@ using mlx-lm version **0.31.2**.
 # About this quant
-mxfp4, but with a bf16 head layer.
 # Original Model Card
@@ -584,4 +588,4 @@ fsdp_config:
 </div>
 </body>
-</html>

 # About this quant
+Just a normal mxfp4. Before the latest version, this model used a bf16 `lm_head` layer, but that has been replaced with a full mxfp4 across the board.
+There are issues with doing any sort of hybrid quantization using mxfp4/mxfp8/nvfp4, and the patches I previously put in were good enough to create a model, but not good enough to run them properly. Big thanks to @circadesu for bringing these issues to my attention.
+For now, I've removed the bf16 head layer and fallen back to a "normal" quant for both mxfp4 and nvfp4. I'm also deleting the mxfp4-mxfp8 and nvfp4-mxfp8 quants of this model entirely. I want to revisit this, but it will take substantial work that I'm not ready to take on quite yet.
 # Original Model Card
 </div>
 </body>
+</html>

config.json CHANGED Viewed

@@ -42,16 +42,9 @@
         "mode": "mxfp4"
     },
     "rms_norm_eps": 1e-05,
-    "rope_parameters": {
-        "partial_rotary_factor": 0.5,
-        "rope_theta": 1000000,
-        "rope_type": "default"
-    },
-    "rope_scaling": {
-        "type": "dynamic",
-        "factor": 1.0
-    },
-    "rope_theta": 10000.0,
     "routed_scaling_factor": 1.0,
     "tie_word_embeddings": false,
     "topk_group": 1,
@@ -59,4 +52,4 @@
     "use_cache": true,
     "use_qk_norm": false,
     "vocab_size": 151552
-}

         "mode": "mxfp4"
     },
     "rms_norm_eps": 1e-05,
+    "rope_type": "default",
+    "rope_scaling": null,
+    "rope_theta": 1000000.0,
     "routed_scaling_factor": 1.0,
     "tie_word_embeddings": false,
     "topk_group": 1,
     "use_cache": true,
     "use_qk_norm": false,
     "vocab_size": 151552
+}

model-00001-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:14c03ae3c1f3fd957eba9d12cfc64040932f75cfa7d353abca49da5a6bb1d550
+size 5035673775

model-00002-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:43d0acd45153e1772e0f06d19d73d14f4a10a7c751305d513b5e664b91658827
+size 5347997371

model-00003-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:a704a71bf75e8a026dda31347bc79d889021975041a612f52871a3f57f7becdd
+size 5070188936

model-00004-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:05088b3f7d0a4f009863ac794fc9b5bbc85c482f031b0ee3c1fff4ee9cfb09a6
+size 5347997498

model-00005-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:5a0c8dbfbd2443683e4fc8b91396eb895434ec7cf86dfc7156419c55ba6d1e94
+size 5001967389

model-00006-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:08d848c38a116f714cee1791e1c6b67a681ef2f048177f5141f17252033a8a37
+size 5347997488

model-00007-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:9250fb61918569dcb44453c46133e5e2f5cd7367308c73faabf5d426b0716142
+size 5001967351

model-00008-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:8f3adb99048a80337c783520f0f760d8d6e2148f283bd192ab6bda4c711ce326
+size 5347997482

model-00009-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:fb19c366f0a36977d135585e0cdccae6805338df955897eea55510d27f62dac1
+size 5070188990

model-00010-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:7777af1616da4c3e04041119959b0111ecf77e7f5e88d757a6d5687051e94398
+size 5347997478

model-00011-of-00011.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:a3cac3df9afb39885d45ef8e304eb124b27a7afd3097bd748c79e75fbc6cf7ee
+size 4881621573

model.safetensors.index.json CHANGED Viewed

The diff for this file is too large to render. See raw diff