Instructions to use philbert440/Qwen3.8-27B-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use philbert440/Qwen3.8-27B-NVFP4 with Transformers:

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

pipe = pipeline("image-text-to-text", model="philbert440/Qwen3.8-27B-NVFP4")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("philbert440/Qwen3.8-27B-NVFP4")
model = AutoModelForMultimodalLM.from_pretrained("philbert440/Qwen3.8-27B-NVFP4", 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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use philbert440/Qwen3.8-27B-NVFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "philbert440/Qwen3.8-27B-NVFP4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "philbert440/Qwen3.8-27B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/philbert440/Qwen3.8-27B-NVFP4

SGLang

How to use philbert440/Qwen3.8-27B-NVFP4 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 "philbert440/Qwen3.8-27B-NVFP4" \
    --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": "philbert440/Qwen3.8-27B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "philbert440/Qwen3.8-27B-NVFP4" \
        --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": "philbert440/Qwen3.8-27B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use philbert440/Qwen3.8-27B-NVFP4 with Docker Model Runner:
```
docker model run hf.co/philbert440/Qwen3.8-27B-NVFP4
```

philbert440 commited on Aug 14

Commit

c36af8e

verified ·

1 Parent(s): df36d26

Model card: V100 measured performance (warm multi-pass), draft-mode + cudagraph sizing guidance, chart

Browse files

Files changed (2) hide show

README.md +34 -4
images/throughput-v100.png +0 -0

README.md CHANGED Viewed

@@ -53,9 +53,38 @@ ModelOpt-exported NVFP4 checkpoints require capability 7.5+ and reject Volta. On
 ## Measured performance
-Benchmarks on 2×V100-32GB (1Cat-vLLM 1.2.2, TP2, `fp8_e5m2` KV, official Qwen3.8 thinking-mode
-sampling) are being run now — multi-pass, warm-serve methodology, MTP K=2 in both draft modes.
-This section will be updated with the results and a throughput chart.
 ## The base model
@@ -122,7 +151,8 @@ python -m vllm.entrypoints.openai.api_server \
   --tensor-parallel-size 2 --gpu-memory-utilization 0.78 \
   --max-model-len 32768 --kv-cache-dtype fp8_e5m2 \
   --enable-prefix-caching --reasoning-parser qwen3 \
-  --speculative-config '{"method":"mtp","num_speculative_tokens":2,"attention_backend":"FLASH_ATTN_V100","draft_sample_method":"greedy"}'
 ```
 SM70 notes, learned the hard way:

 ## Measured performance
+![MTP draft-mode throughput](images/throughput-v100.png)
+**Methodology:** warm serve (2 discarded warmup generations), fixed-length generations via
+`ignore_eos` so every run produces exactly the stated token count, official Qwen3.8 sampling per
+mode (thinking `1.0/0.95/20`, instruct `0.7/0.80/20 + presence 1.5`), varied prompts. Reported
+as mean ± sd tokens/s. Rig: 2×V100-32GB, 1Cat-vLLM 1.2.2, TP2, `fp8_e5m2` KV,
+`max_num_seqs 4`, MTP K=2.
+| Regime | greedy draft | probabilistic draft |
+|---|---|---|
+| 512-tok, thinking (n=10) | 53.0 ± 2.5 | **54.4 ± 1.3** |
+| 2048-tok, thinking (n=3) | 51.1 ± 2.5 | **54.1 ± 1.1** |
+| 512-tok, instruct (n=6) | 50.5 ± 1.8 | 50.0 ± 1.6 |
+| Mean acceptance length, whole workload | 2.25 | 2.46 |
+**Concurrency** (4-way, 512-tok, aggregate): **~170 tok/s** with
+`{"cudagraph_mode":"piecewise"}` (auto capture sizes), ~165 with `full_and_piecewise` — on par
+with the [W4A16 sibling](https://huggingface.co/philbert440/Qwen3.8-27B-W4A16-AWQ). One
+sizing rule matters: with MTP, each sequence schedules `K+1` tokens per step, so **never set
+explicit `cudagraph_capture_sizes` below `max_num_seqs × (K+1)`** — a cap of `[1,2,4,8]` at
+batch 4 pushes concurrent decode off CUDA graphs and collapses aggregate throughput ~3×
+(measured 55–72 tok/s; reproduces identically on the W4A16 sibling, so it's a config trap, not
+a format property). Engine-default auto sizing is correct.
+**Pick the draft mode by workload:** verification rejection-samples against the target model,
+so output quality is identical either way. At the official temp-1.0 thinking sampling,
+**probabilistic** matches the verified distribution and wins (+2–6%); on low-temperature
+workloads the two converge (see instruct row); at `temperature 0` greedy is the natural choice.
+Quality validation (passed on this rig): factual coherence, think-tag discipline (zero
+`<think>` leakage with thinking disabled), vision (image understanding through the VLM path),
+GSM8K sample 3/3, and long-form generation with no repetition/degeneration.
 ## The base model
   --tensor-parallel-size 2 --gpu-memory-utilization 0.78 \
   --max-model-len 32768 --kv-cache-dtype fp8_e5m2 \
   --enable-prefix-caching --reasoning-parser qwen3 \
+  --compilation-config '{"cudagraph_mode":"piecewise"}' \
+  --speculative-config '{"method":"mtp","num_speculative_tokens":2,"attention_backend":"FLASH_ATTN_V100","draft_sample_method":"probabilistic"}'
 ```
 SM70 notes, learned the hard way:

images/throughput-v100.png ADDED Viewed