Instructions to use mconcat/GLM-5.1-FP8-Dynamic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mconcat/GLM-5.1-FP8-Dynamic with Transformers:

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

pipe = pipeline("text-generation", model="mconcat/GLM-5.1-FP8-Dynamic")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mconcat/GLM-5.1-FP8-Dynamic")
model = AutoModelForCausalLM.from_pretrained("mconcat/GLM-5.1-FP8-Dynamic", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mconcat/GLM-5.1-FP8-Dynamic with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mconcat/GLM-5.1-FP8-Dynamic"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mconcat/GLM-5.1-FP8-Dynamic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mconcat/GLM-5.1-FP8-Dynamic

SGLang

How to use mconcat/GLM-5.1-FP8-Dynamic 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 "mconcat/GLM-5.1-FP8-Dynamic" \
    --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": "mconcat/GLM-5.1-FP8-Dynamic",
		"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 "mconcat/GLM-5.1-FP8-Dynamic" \
        --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": "mconcat/GLM-5.1-FP8-Dynamic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mconcat/GLM-5.1-FP8-Dynamic with Docker Model Runner:
```
docker model run hf.co/mconcat/GLM-5.1-FP8-Dynamic
```

mconcat commited on Apr 10

Commit

9ef68c3

verified ·

1 Parent(s): 40b97bf

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +22 -2

README.md CHANGED Viewed

@@ -83,12 +83,32 @@ vllm serve mconcat/GLM-5.1-FP8-Dynamic \
 ## Notes
-- This is a 754B MoE model (~40B active per token). Requires multi-GPU setup for inference.
 - FP8 E4M3 provides ~2x compression over BF16 with minimal quality degradation.
-- Compatible with Hopper (SM90) and Blackwell (SM100+) GPUs.
 - Dynamic activation scaling — scales computed at inference time, not baked into the checkpoint.
 - GLM-5.1 does not ship MTP weights despite `num_nextn_predict_layers=1` in config.
 ## Quantization Process
 - **Tool**: Custom layer-by-layer pipeline with native `torch.float8_e4m3fn` dtype

 ## Notes
+- This is a 754B MoE model (~40B active per token). Requires multi-GPU setup for inference (8x 80GB+ GPUs recommended).
 - FP8 E4M3 provides ~2x compression over BF16 with minimal quality degradation.
+- Compatible with Hopper (SM90) and Blackwell GPUs.
 - Dynamic activation scaling — scales computed at inference time, not baked into the checkpoint.
 - GLM-5.1 does not ship MTP weights despite `num_nextn_predict_layers=1` in config.
+## Blackwell SM120 Patch (RTX PRO 6000 / Workstation GPUs)
+If running on Blackwell workstation GPUs (SM 12.0), vLLM 0.19.0 requires patches for FlashMLA sparse attention support:
+```bash
+# Patch 1: FlashMLA ops - add SM120 to sparse support check
+FLASHMLA_OPS=$(python -c "import vllm, os; print(os.path.join(os.path.dirname(vllm.__file__), 'v1/attention/ops/flashmla.py'))") && \
+sed -i 's/is_device_capability_family(90)\s*or current_platform.is_device_capability_family(100)/is_device_capability_family(90) or current_platform.is_device_capability_family(100) or current_platform.is_device_capability_family(120)/' "$FLASHMLA_OPS"
+# Patch 2: FlashMLA sparse backend - add SM12 to capability check
+FLASHMLA_SPARSE=$(python -c "import vllm, os; print(os.path.join(os.path.dirname(vllm.__file__), 'v1/attention/backends/mla/flashmla_sparse.py'))") && \
+sed -i 's/return capability.major in \[9, 10\]/return capability.major in [9, 10, 12]/' "$FLASHMLA_SPARSE"
+# Patch 3: FlashMLA dense backend (if exists)
+FLASHMLA_DENSE=$(python -c "import vllm, os; print(os.path.join(os.path.dirname(vllm.__file__), 'v1/attention/backends/mla/flashmla.py'))") && \
+sed -i 's/return capability.major in \[9, 10\]/return capability.major in [9, 10, 12]/' "$FLASHMLA_DENSE" 2>/dev/null || true
+```
+These patches add SM120 (Blackwell workstation) to the supported compute capability list for GLM-5.1's DSA sparse attention.
 ## Quantization Process
 - **Tool**: Custom layer-by-layer pipeline with native `torch.float8_e4m3fn` dtype