Instructions to use sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP with Transformers:

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

pipe = pipeline("text-generation", model="sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP")
model = AutoModelForMultimodalLM.from_pretrained("sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP

SGLang

How to use sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP 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 "sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP" \
    --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": "sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP",
		"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 "sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP" \
        --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": "sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP with Docker Model Runner:
```
docker model run hf.co/sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP
```

sakamakismile commited on Apr 26

Commit

257c4c7

verified ·

1 Parent(s): 12761b7

README: n=3 spec recommended (132/105/106 tok/s); document mtp.fc bf16 = Lorbus trick

Browse files

Files changed (1) hide show

README.md +19 -2

README.md CHANGED Viewed

@@ -73,11 +73,13 @@ vllm serve sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP \
     --trust-remote-code \
     --gpu-memory-utilization 0.85 \
     --max-model-len 8192 \
     --quantization modelopt \
-    --speculative-config '{"method":"qwen3_5_mtp","num_speculative_tokens":1}'
 ```
-`num_speculative_tokens: 1` because the model has a single MTP layer (`mtp_num_hidden_layers=1`). The `qwen3_5_mtp` method handler is what vLLM uses for the Qwen3.5/3.6 family (internal `model_type: qwen3_5`); plain `mtp` also works as a fallback.
 ### Without speculation
@@ -86,9 +88,24 @@ vllm serve sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP \
     --trust-remote-code \
     --gpu-memory-utilization 0.85 \
     --max-model-len 8192 \
     --quantization modelopt
 ```
 ## Hardware target
 Built and tested on **NVIDIA RTX PRO 6000 Blackwell (SM120)**. Should also work on **RTX 5090** and other Blackwell consumer/workstation cards with sufficient VRAM (the model is roughly 14 GB after NVFP4 + ~850 MB of bf16 MTP/conv1d/lm_head).

     --trust-remote-code \
     --gpu-memory-utilization 0.85 \
     --max-model-len 8192 \
+    --language-model-only \
     --quantization modelopt \
+    --reasoning-parser qwen3 \
+    --speculative-config '{"method":"qwen3_5_mtp","num_speculative_tokens":3}'
 ```
+`num_speculative_tokens: 3` is the **recommended setting** even though the model has a single MTP layer (`mtp_num_hidden_layers=1`): vLLM applies that one layer recursively three times per draft pass, and on Qwen3.5/3.6-family models the per-position acceptance rates stay high enough (typically ~87 / ~72 / ~61 % at positions 1 / 2 / 3) that mean accepted-length lands around 3.0 — which is what unlocks 100+ tok/s on a single Blackwell card. `num_speculative_tokens: 1` (the value originally documented for this family) is a safer fallback if you hit a draft-path bug. The `qwen3_5_mtp` method handler is what vLLM uses for the Qwen3.5/3.6 family (internal `model_type: qwen3_5`); plain `mtp` also works as a fallback.
 ### Without speculation
     --trust-remote-code \
     --gpu-memory-utilization 0.85 \
     --max-model-len 8192 \
+    --language-model-only \
     --quantization modelopt
 ```
+## Verified throughput
+Single-request decode, T = 0, 9 runs across 3 prompt lengths on 1 × RTX PRO 6000 Blackwell, vLLM 0.19.1rc1:
+| Prompt | Tokens | n=1 tok/s | **n=3 tok/s** |
+|---|---|---|---|
+| Short (50 tok) | 50 | ~71 | **132.5** |
+| Medium (350 tok) | 350 | ~85 | **105.5** |
+| Long-form (700 tok) | 700 | ~85 | **106.5** |
+GPU memory at load: ~15 GB. Mean acceptance length 1.93 / 2.0 at n=1, ~3.0 / 4.0 at n=3 (per-position accept ~87 / 72 / 61 %, matches Pulsate1680's RTX PRO 4500 Blackwell result on this same checkpoint).
+The `mtp.fc` weight is kept in **bf16** in the safetensors (not NVFP4) — equivalent to the Lorbus-style "dequantize the fusion layer in the file" trick, applied to NVFP4 instead of AutoRound. This is a side effect of the `*mtp*` ignore entry in the modelopt config, but it is the load-bearing detail behind the n=3 throughput.
 ## Hardware target
 Built and tested on **NVIDIA RTX PRO 6000 Blackwell (SM120)**. Should also work on **RTX 5090** and other Blackwell consumer/workstation cards with sufficient VRAM (the model is roughly 14 GB after NVFP4 + ~850 MB of bf16 MTP/conv1d/lm_head).