Instructions to use VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT with Transformers:

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

pipe = pipeline("image-text-to-text", model="VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT")
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("VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT")
model = AutoModelForMultimodalLM.from_pretrained("VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT", 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 VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT",
		"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/VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT

SGLang

How to use VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT 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 "VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT" \
    --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": "VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT",
		"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 "VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT" \
        --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": "VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT",
		"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 VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT with Docker Model Runner:
```
docker model run hf.co/VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT
```

You need to agree to share your Hugging Face information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

CVC-Qwen3.5-35B-A3B-Instruct

Vision-language instruct model finetuned from Qwen/Qwen3.5-35B-A3B-Base.

Mixture of experts: 35B total parameters, ~3B active per token (256 experts, top-8), 40 decoder layers, plus a 27-layer vision encoder. Weights are bf16.

Training

Two supervised finetuning stages, both at sequence length 16384 on 64 nodes of JUPITER (4x GH200 per node):

Instruct SFT — one pass over the LLaVA-OneVision-1.5-Instruct + Nemotron-VLM-V2 (197 sources), 2274 steps, 16.7B tokens
Decay anneal — a further 326 steps over the same sources reweighted towards document and OCR data, with the learning rate decayed to zero. ~2B tokens.

Benchmark Results

Evaluated with lmms-eval + vLLM on JUPITER (8 nodes, 32x GH200).

Benchmark	CVC-35B-A3B-Instruct	Qwen3.5-35B-A3B-Base	Qwen3.5-35B-A3B
MMU_val (acc)	0.6833	0.6056	0.6267
MME (perception)	1769.7	1666.4	1560.8
MME (cognition)	762.9	740.4	751.8
AI2D (exact_match)	0.8598	0.0929	0.6758
OCRBench (accuracy)	0.8760	0.8920	0.9120
MMStar (average)	0.7373	0.7381	0.7289
ChartQA (relaxed_overall)	0.7720	0.2620	0.7572
MMLongBenchDoc (acc)	0.1790	0.2000	0.1980
MMLongBenchDoc (f1)	0.0360	0.0110	0.0060
DocVQA_val (anls)	0.9130	0.6515	0.9539
TextVQA_val (exact_match)	0.6573	0.2669	0.7665
InfoVQA_val (anls)	0.8570	0.3412	0.8622
SeedBench (seed_all)	0.7429	0.7479	0.7152
RealWorldQA (exact_match)	0.7438	0.6288	0.7739

Training data mixture

Both stages draw from the same 197-source LLaVA-OneVision-1.5-Instruct with Nemotron-VLM-Dataset-V2. The two stages differ only in how that blend is sampled. The exact list of subsets can be found on DATA_MIX.md.

Stable (instruct SFT) blend_epochized, every sample seen once (one deterministic full pass, ~16.7B tokens total). Composition follows each source's natural size:

Category	Share of stable-phase tokens	Example subsets
OCR / scene-text (core)	1.34%	`textvqa`, `st_vqa`, `textocr_gpt4v`, `ocrvqa`, `llavar`, `textcaps`, `sroie_data`
Document VQA (core)	0.31%	`docvqa_train`, `infographic_vqa`
Chart / plot (core)	8.89%	`chartqa`, `FigureQA`, `dvqa`, `chart2text`, `tqa`, `plotqa_*`
Spatial / real-world (core)	1.90%	`gqa`, `aokvqa`, `visual7w`, `vsr`, `CLEVR`, `CLEVR-Math`, `vg`
Text-only reasoning/math/code	30.18%	`orca_agentinstruct`, `openmathinstruct`, `magpie_ultra`, `magpie_pro`, `open_orca`
General captioning/instruct	21.82%	`llava_instruct`, `sharegpt4o`, `sharegpt4v`, `allava`, `cambrian`, `Docmatix`, `svit`, `wit`
Nemotron-v2 text	17.72%	`wiki_*`x10, `sparsetables`, `nights_cot`
Document/OCR-adjacent (not core)	14.70%	`ureader_*`, `allenai_pixmo_docs`, `tinychart_train`, `robut_wikisql/wtq/sqa`, `mapqa`, `vistext`
Agentic/web	1.60%	`VisualWebInstruct`, `alfworld`, `rootsautomation`, `sherlock`
Science/diagram reasoning	1.00%	`ai2d`, `geo170k_*`, `Geometry3K`, `scienceqa`, `PMC-VQA`
Spatial/counting extras (not core)	0.53%	`tallyqa`, `raven`, `Super-CLEVR`, `IconQA`, `VizWiz`
Video (tiny, line-count fallback)	~0.00%	`perception_test_1/2/cot`, `breakfast_actions`, `ego_exo_learn`

Decay (anneal) switched to a weighted blend (continuous sampling, ~2B-token budget) with four target categories deliberately upweighted:

Category	Share of decay-phase tokens
OCR / scene-text (core, upweighted)	25%
Document VQA (core, upweighted)	15%
Spatial / real-world (core, upweighted)	12%
Chart / plot (core, upweighted)	8%
Text-only reasoning/math/code (replay)	13.79%
General captioning/instruct (replay)	9.97%
Nemotron-v2 text (replay)	8.10%
Document/OCR-adjacent, not core (replay)	6.72%
Agentic/web (replay)	0.73%
Science/diagram reasoning (replay)	0.46%
Spatial/counting extras, not core (replay)	0.24%
Video, tiny fallback (replay)	~0.00%

Notes

Uses the Qwen3.5 chat template. Unlike the upstream instruct template, this one renders <think> blocks on every assistant turn rather than only the last, matching how the model was trained.
eos_token is <|im_end|>, and generation_config.json stops on both <|im_end|> (248046) and <|endoftext|> (248044).

Downloads last month: 1

Safetensors

Model size

35B params

Tensor type

BF16

Model tree for VLR-CVC/CVC-Qwen3.5-35B-A3B-SFT

Base model

Qwen/Qwen3.5-35B-A3B-Base

Finetuned

(98)

this model

Finetunes

1 model