Instructions to use axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic") 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("axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic") model = AutoModelForMultimodalLM.from_pretrained("axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic", 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 axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "axteam/MiniCPM-V-4.7-35B-A3B-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": "axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic", "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/axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic
- SGLang
How to use axteam/MiniCPM-V-4.7-35B-A3B-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 "axteam/MiniCPM-V-4.7-35B-A3B-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": "axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic", "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 "axteam/MiniCPM-V-4.7-35B-A3B-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": "axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic", "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 axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic with Docker Model Runner:
docker model run hf.co/axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic
MiniCPM-V-4.7-35B-A3B ยท FP8 (dynamic)
8-bit quantization of openbmb/MiniCPM-V-4.7-35B-A3B for vLLM serving. Quantized by axteam.
Revision 2026-11-02 (config fix): the earlier
config.jsonexcluded the vision tower from quantization with the patternre:model\.vision_tower\..*, which does not match the names vLLM uses for these modules (vpm.*,vit_merger.*). vLLM 0.32 therefore built the vision layers as FP8 and loaded the BF16 vision weights into them, so image input gave wrong answers; text input was not affected. Onlyconfig.jsonchanged (vLLM's names were added to the ignore list asre:vpm\..*andre:vit_merger\..*, and the existing patternsre:model\.visual\..*,re:model\.vision_tower\..*andre:model\.merger\..*were widened tore:(model\.)?visual\..*,re:(model\.)?vision_tower\..*andre:(model\.)?merger\..*); the weights are the same. Checked with vLLM 0.32.0 on 2026-11-02: the vision weights now load as BF16, and a synthetic test image (red circle, blue square, the caption "QWEN TEST 2026") is described correctly (with vLLM 0.32.0 + transformers 5.21 the image preprocessing fails, see Requirements; for this check only, the image processor's missingdownsample_modeargument was filled in with its default). If you downloaded this repo before 2026-11-01 23:41:44 UTC (revisiona1f77bb), downloadconfig.jsonagain.
| Base model | openbmb/MiniCPM-V-4.7-35B-A3B |
| Method | FP8 (dynamic) |
| Format | compressed-tensors float-quantized (scheme FP8_DYNAMIC) |
| Weights / activations | W8A8 โ FP8 E4M3 weights (per-channel), dynamic per-token FP8 activations |
| Architecture | MoE (Qwen3.5-MoE backbone) ยท vision ยท multimodal |
| Size on disk | ~36.9 GB |
| Calibration | None (data-free) |
| License | Apache-2.0 |
Quantization details
- Scheme โ W8A8 โ FP8 E4M3 weights (per-channel), dynamic per-token FP8 activations, stored as compressed-tensors
float-quantized(schemeFP8_DYNAMIC). - Kept in BF16 โ the vision tower & vision-text merger, token embeddings, the LM head, all RMSNorms, the MoE routers (
mlp.gate,shared_expert_gate) โ about 5% of all parameters in this build. Quantized: the attention (incl. linear-attention) and MoE/MLP projections. - Post-training quantization (no retraining). Tokenizer /
chat_templateidentical to the original.
Benchmarks (measured)
Every column was measured with the same harness and settings โ lm-evaluation-harness on vLLM, greedy, completion-style prompts without a chat template (0- or 5-shot, as listed in each row). MMLU here is lm-eval's generative variant (mmlu_generative: the model writes the answer letter and only an exact match on the first line counts), so it is not directly comparable with MMLU scores measured another way (e.g. log-likelihood scoring). Percentages are recovery vs. the BF16 original.
| Benchmark | BF16 (original) | FP8-dynamic (this repo) | AWQ-8bit | NVFP4A16 | NVFP4 | AWQ-4bit |
|---|---|---|---|---|---|---|
| MMLU (5-shot, generative) | 85.65 | 85.61 (100.0%) | 85.71 (100.1%) | 85.76 (100.1%) | 85.41 (99.7%) | 85.31 (99.6%) |
| GSM8K (5-shot, strict) | 90.14 | 89.69 (99.5%) | 89.23 (99.0%) | 90.75 (100.7%) | 90.22 (100.1%) | 90.52 (100.4%) |
| ARC-Challenge (0-shot) | 60.92 | 61.01 (100.1%) | 61.18 (100.4%) | 60.84 (99.9%) | 61.09 (100.3%) | 62.03 (101.8%) |
| Winogrande (5-shot) | 80.11 | 79.87 (99.7%) | 78.93 (98.5%) | 79.32 (99.0%) | 77.11 (96.3%) | 79.56 (99.3%) |
| KMMLU (5-shot, direct) | 64.95 | 64.77 (99.7%) | 64.78 (99.7%) | 64.65 (99.5%) | 64.38 (99.1%) | 64.67 (99.6%) |
| HAE-RAE (0-shot) | 73.14 | 73.69 (100.8%) | 73.42 (100.4%) | 73.51 (100.5%) | 74.06 (101.3%) | 72.69 (99.4%) |
| Average | 75.82 | 75.77 (99.9%) | 75.54 (99.6%) | 75.80 (100.0%) | 75.38 (99.4%) | 75.80 (100.0%) |
Evaluation (measured)
Fidelity to the BF16 original on an internal mixed English/Korean set (100 prompts: the BF16 original's greedy answers, scored under each build with the same prompts and scoring code). This build was measured with vLLM 0.32 on 2026-11-01; the BF16 reference column was measured on 2026-11-01.
| Metric | This model | BF16 original |
|---|---|---|
| Perplexity (lower = closer) | 1.2269 | 1.2063 (+1.7%) |
| Top-1 next-token agreement | 96.0% | 98.1% |
These measure fidelity to the base model, not absolute task scores.
Requirements
| Runtime | vLLM โ the architecture is in vLLM โฅ 0.32.0; tested here with vLLM 0.32.0. vLLM 0.32 needs transformers 5.18.x or 5.19.x for this model (5.17 does not know minicpmv4_7); our runs used transformers 5.21.0 installed with --no-deps, text-only โ image input fails in that setup (vLLM's MiniCPM-V image processor vs transformers 5.21); only the 2026-11-02 image check in the revision note above ran images, with the processor's missing downsample_mode argument filled in for the test. |
| vLLM install | vLLM 0.32 itself needs Python 3.11โ3.14, torch 2.13 (CUDA 13.0 build, so an NVIDIA driver with CUDA 13 support), transformers โฅ5.16.1,<5.20 and compressed-tensors 0.17.0 (both pinned by vLLM; 0.17.0 reads this checkpoint, which was saved with 0.19.0 โ do not upgrade it separately). |
| GPU | NVIDIA Turing (sm75, e.g. T4) or newer. FP8 math runs on Ada / Hopper / Blackwell (sm89+); on Turing and Ampere vLLM switches to weight-only FP8 automatically. V100 (sm70) is not supported. (Kernel choice per GPU from vLLM 0.32's selection code; only B200 was run here.) |
| Memory | Weights: 36.9 GB. Add KV cache (grows with context length ร concurrent requests) and a few GB for activations / CUDA graphs; use --tensor-parallel-size N when the weights do not fit one GPU. |
| Transformers (without vLLM) | transformers>=5.18.0 (FP8 checked). |
| trust_remote_code | not needed |
Usage (vLLM)
Compatibility (checked 2026-10-31): with vLLM 0.32.0 and Transformers 5.17.0 the server did not start โ Transformers does not recognise model type
minicpmv4_7and this repo ships no custom modeling code. Use a Transformers / vLLM release that supports MiniCPM-V 4.7.
Start an OpenAI-compatible server:
vllm serve axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic --max-model-len 32768
Query it:
curl http://localhost:8000/v1/chat/completions -H "Content-Type: application/json" -d '{
"model": "axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic",
"messages": [{"role": "user", "content": "๋ํ๋ฏผ๊ตญ์ ์๋๋ ์ด๋์ธ๊ฐ์?"}],
"temperature": 1.0, "top_p": 0.95, "top_k": 20, "max_tokens": 512
}'
Offline inference in Python:
from vllm import LLM, SamplingParams
llm = LLM(model="axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic", max_model_len=32768)
params = SamplingParams(temperature=1.0, top_p=0.95, top_k=20, max_tokens=1024)
out = llm.chat([{"role": "user", "content": "Explain model quantization in two sentences."}], params)
print(out[0].outputs[0].text)
# image input (URL or data: URI)
messages = [{"role": "user", "content": [
{"type": "image_url", "image_url": {"url": "https://example.com/your-image.png"}},
{"type": "text", "text": "Describe this image."},
]}]
out = llm.chat(messages, params)
print(out[0].outputs[0].text)
Recommended sampling (from the original
generation_config.json):temperature=1.0,top_p=0.95,top_k=20. Thinking mode is off by default in this model's chat template; to turn it on passchat_template_kwargs={"enable_thinking": True}(Pythonllm.chat(..., chat_template_kwargs=...), or"chat_template_kwargs": {"enable_thinking": true}in the request body). Text-only serving: add--limit-mm-per-prompt '{"image": 0, "video": 0}'to skip loading the vision encoder's memory budget.
Reducing KV-cache memory (optional, serve-time)
The KV cache is kept in BF16 by default. KV-cache precision is a runtime choice, independent of the weight quantization above. This is a hybrid model: only its 10 full-attention layers keep a KV cache, while its 30 linear-attention (Gated DeltaNet) layers keep a fixed-size state per request, which --kv-cache-dtype does not change. FP8 halves the full-attention KV cache, so the saving grows with context length and is small for many short requests:
vllm serve axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic --kv-cache-dtype fp8
License
Quantized derivative of openbmb/MiniCPM-V-4.7-35B-A3B (Apache-2.0). Original weights & tokenizer belong to OpenBMB.
ํ๊ตญ์ด ์์ฝ
openbmb/MiniCPM-V-4.7-35B-A3B์ FP8 (W8A8) ์์ํ ๋น๋ ยท ์ ์ axteam
๊ฐ์
| ํญ๋ชฉ | ๋ด์ฉ |
|---|---|
| ์๋ณธ ๋ชจ๋ธ | openbmb/MiniCPM-V-4.7-35B-A3B |
| ์์ํ ๋ฐฉ์ | FP8 ๋์ ์์ํ (W8A8) |
| ์ ์ฅ ํ์ | compressed-tensors float-quantized (scheme FP8_DYNAMIC) |
| ๋ชจ๋ธ ๊ตฌ์กฐ | MoE (Qwen3.5-MoE ๋ฐฑ๋ณธ) ยท ๋ฉํฐ๋ชจ๋ฌ(๋น์ ) |
| ์ฉ๋ | ์ฝ 36.9 GB |
| ๋ณด์ ๋ฐ์ดํฐ | ์์ (๊ฐ์ค์น ์ค์ผ์ผ์ ๊ฐ์ค์น๋ง์ผ๋ก ๊ณ์ฐ, ํ์ฑํ ์ค์ผ์ผ์ ์คํ ์ ๊ณ์ฐ) |
์์ํ ์์ธ
- ๊ฐ์ค์น: FP8(E4M3), ์ถ๋ ฅ ์ฑ๋๋ณ ์ค์ผ์ผ
- ํ์ฑํ: ์คํ ์ ํ ํฐ๋ณ ๋์ FP8
- ์์ํ ์ ์ฐจ: PTQ(ํ์ต ํ ์์ํ, ์ฌํ์ต ์์), ๊ฐ์ค์น ์ค์ผ์ผ์ ๋ฏธ๋ฆฌ ๊ณ์ฐํ๊ณ ํ์ฑํ ์ค์ผ์ผ์ ์คํ ์ ํ ํฐ๋ง๋ค ๊ณ์ฐ
- BF16 ์ ์ง: ๋น์ ํ์, ๋น์ -ํ
์คํธ ํฌ์์ธต(merger), ํ ํฐ ์๋ฒ ๋ฉ, LM ํค๋, ๋ชจ๋ RMSNorm, MoE ๋ผ์ฐํฐ(
mlp.gate)์ ๊ณต์ ์ ๋ฌธ๊ฐ ๊ฒ์ดํธ(shared_expert_gate) โ ์ ์ฒด ํ๋ผ๋ฏธํฐ์ ์ฝ 5% - ์์ํ ๋์: ์ดํ ์ (์ ํ ์ดํ ์ ํฌํจ)๊ณผ MoE/MLP์ ์ ํ์ธต(projection)
- ์๋ณธ ํ์ผ: ํ ํฌ๋์ด์ ยทchat template์ ์๋ณธ๊ณผ ๋์ผ
ํ์ง (์ค์ธก)
- ์๋ณธ ๋๋น ์ถฉ์ค๋: PPL 1.2269 (์๋ณธ 1.2063, +1.7%, ๋ฎ์์๋ก ์๋ณธ์ ๊ฐ๊น์), ๋ค์ ํ ํฐ top-1 ์ผ์น์จ 96.0% (์๋ณธ ์์ฒด ๊ธฐ์ค๊ฐ 98.1%) โ ๋ด๋ถ ์์ดยทํ๊ตญ์ด ํผํฉ ํ๋กฌํํธ 100๊ฐ, ์๋ณธ(BF16)์ greedy ๋ต๋ณ์ ๊ธฐ์ค ํ ์คํธ๋ก ์ธก์
- ๋ฒค์น๋งํฌ ํ๊ท : 75.77 (์๋ณธ 75.82, 99.9%) โ MMLU(์์ฑํ)ยทGSM8KยทARC-ChallengeยทWinograndeยทKMMLUยทHAE-RAE 6์ข ํ๊ท , lm-evaluation-harness + vLLM, chat template ์๋ ์์ฑํ ํ๋กฌํํธ(๊ณผ์ ๋ณ 0ยท5-shot, ํ ์ฐธ๊ณ )ยทgreedy, ๋ชจ๋ ๋น๋๋ฅผ ๊ฐ์ ์ค์ ์ผ๋ก ์ธก์
- ํ๊ตญ์ด ๋ฒค์น๋งํฌ: KMMLU 64.77 (์๋ณธ 64.95, 99.7%) ยท HAE-RAE 73.69 (์๋ณธ 73.14, 100.8%)
- MMLU ๋ฐฉ์: lm-eval ์์ฑํ ๋ณํ(
mmlu_generative) โ ๋ชจ๋ธ์ด ๋ต ๊ธ์๋ฅผ ์์ฑํ๊ณ ์ฒซ ์ค์ด ์ ํํ ์ผ์นํด์ผ ์ ๋ต์ผ๋ก ์ฑ์ ํ๋ฏ๋ก, ๋ค๋ฅธ ๋ฐฉ์(์: log-likelihood ์ฑ์ )์ผ๋ก ์ฐ MMLU ์ ์์ ์ง์ ๋น๊ต ๋ถ๊ฐ
์๊ตฌ์ฌํญ
- ์คํ ์์ง: vLLM 0.32.0 ์ด์(์ด ๊ตฌ์กฐ๋ฅผ ์ง์ํ๋ ์ฒซ ๋ฒ์ ), ์ด ๋น๋๋ vLLM 0.32.0์์ ์คํ ํ์ธ
- Transformers ๋ฒ์ : vLLM 0.32์ ํจ๊ป ์ฐ๋ ค๋ฉด transformers 5.18.x ๋๋ 5.19.x ํ์ (5.17์
minicpmv4_7๋ฏธ์ง์), ์คํ ํ์ธ์ transformers 5.21.0์--no-deps๋ก ์ค์นํด ํ ์คํธ ์ ๋ ฅ์ผ๋ก ์งํ โ ์ด ์กฐํฉ์์๋ ์ด๋ฏธ์ง ์ ๋ ฅ ์คํจ (vLLM์ MiniCPM-V ์ด๋ฏธ์ง ํ๋ก์ธ์์ transformers 5.21 ๋นํธํ), ์ด๋ฏธ์ง๋ ์ ์์ ์ด๋ ฅ์ 2026-11-02 ์ ๊ฒ๋ง ์ฒ๋ฆฌ๊ธฐ์ ๋น ์งdownsample_mode์ธ์๋ฅผ ์ฑ์ ์คํ - vLLM 0.32 ์ค์น ์กฐ๊ฑด: Python 3.11~3.14, torch 2.13(CUDA 13.0 ๋น๋, CUDA 13 ์ง์ ๋๋ผ์ด๋ฒ ํ์), transformers 5.16.1 ์ด์ 5.20 ๋ฏธ๋ง, compressed-tensors 0.17.0 โ transformersยทcompressed-tensors๋ vLLM์ด ๋ฒ์ ์ ๊ณ ์ ํ๋ฏ๋ก compressed-tensors๋ฅผ ๋ฐ๋ก ์ ๊ทธ๋ ์ด๋ํ์ง ๋ง ๊ฒ (์ด ์ฒดํฌํฌ์ธํธ๋ 0.19.0์ผ๋ก ์ ์ฅ๋์ง๋ง 0.17.0์์๋ ๋ก๋๋จ)
- GPU: NVIDIA Turing(sm75, ์: T4) ์ด์, V100(sm70) ๋ฏธ์ง์ โ FP8 ์ฐ์ฐ์ AdaยทHopperยทBlackwell(sm89 ์ด์)์์ ์ํ, TuringยทAmpere๋ vLLM์ด ๊ฐ์ค์น ์ ์ฉ FP8๋ก ์๋ ์ ํ (GPU๋ณ ์ง์ ์ฌ๋ถ๋ vLLM 0.32 ์ปค๋ ์ ํ ์ฝ๋ ๊ธฐ์ค, ์ค์ ์คํ์ B200์์๋ง ํ์ธ)
- ๋ฉ๋ชจ๋ฆฌ: ๊ฐ์ค์น 36.9 GB + KV ์บ์(์ปจํ
์คํธ ๊ธธ์ด ร ๋์ ์์ฒญ ์์ ๋น๋ก) + ํ์ฑํยทCUDA graph์ฉ ์ GB โ GPU ํ ์ฅ์ ๋ค ๋ค์ด๊ฐ์ง ์์ผ๋ฉด
--tensor-parallel-size N์ผ๋ก ๋ถ์ฐ - Transformers๋ก ์ง์ ์ฌ์ฉ: transformers 5.18.0 ์ด์ (FP8 ๋น๋๋ก ํ์ธ)
- trust_remote_code: ๋ถํ์
์ฌ์ฉ๋ฒ
- ์๋ฒ ์คํ:
vllm serve axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic --max-model-len 32768โ OpenAI ํธํ API - Python:
LLM(model="axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic", max_model_len=32768).chat(...)โ ์ ์ฒด ์์๋ ์ ์์ด ๋ณธ๋ฌธ 'Usage (vLLM)' ์ ์ฐธ๊ณ - ๊ถ์ฅ ์ํ๋ง:
temperature=1.0,top_p=0.95,top_k=20(์๋ณธgeneration_config.json๊ธฐ์ค) - thinking ๋ชจ๋: ๊ธฐ๋ณธ ๊บผ์ง โ ์ผ๋ ค๋ฉด
chat_template_kwargs={"enable_thinking": True}์ ๋ฌ(Python์llm.chat()์ธ์, HTTP๋ ์์ฒญ ๋ณธ๋ฌธ์"chat_template_kwargs": {"enable_thinking": true}) - ์ด๋ฏธ์ง ์
๋ ฅ: ๋ฉ์์ง์
image_urlํญ๋ชฉ ์ฌ์ฉ - ํ
์คํธ๋ง ์ฌ์ฉํ ๋:
--limit-mm-per-prompt '{"image": 0, "video": 0}'โ ๋น์ ํ์์ฉ ๋ฉ๋ชจ๋ฆฌ ์์ฝ์ ์๋ตํด ๋ฉ๋ชจ๋ฆฌ ์ ์ฝ - KV ์บ์ ์ ๊ฐ(์ ํ):
vllm serve์--kv-cache-dtype fp8์ถ๊ฐ โ ํ์ด๋ธ๋ฆฌ๋ ๊ตฌ์กฐ๋ผ KV ์บ์๋ ํ ์ดํ ์ 10๊ฐ ์ธต์๋ง ์๊ณ , ์ ํ ์ดํ ์ (Gated DeltaNet) 30๊ฐ ์ธต์ ์์ฒญ๋ณ ๊ณ ์ ํฌ๊ธฐ ์ํ๋ ์ด ์ต์ ์ผ๋ก ๋ฐ๋์ง ์์ โ ํ ์ดํ ์ KV ์บ์๋ง ์ ๋ฐ์ผ๋ก ์ค์ด ๊ธด ์ปจํ ์คํธ์์ ํจ๊ณผ๊ฐ ํฌ๊ณ ์งง์ ์์ฒญ์ด ๋ง์ ๋๋ ์์ (๊ฐ์ค์น ์์ํ์ ๋ณ๊ฐ์ธ ์คํ ์ต์ )
์ฐธ๊ณ
- ํธํ์ฑ (2026-10-31 ํ์ธ): vLLM 0.32.0 + transformers 5.17.0์
minicpmv4_7์ ์ธ์ํ์ง ๋ชปํด ์๋ฒ ์คํ ์คํจ โ transformers 5.18.xยท5.19.x ํ์ (์ ์๊ตฌ์ฌํญ ์ฐธ๊ณ ) - ์์ ์ด๋ ฅ (2026-11-02, ์ค์ ์์ ): ์ด์
config.json์ ๋น์ ํ์๋ฅผ ์์ํ์์ ๋นผ๋ ๊ท์น(re:model\.vision_tower\..*)์ด vLLM์ด ์ฐ๋ ๋ชจ๋ ์ด๋ฆ(vpm.*,vit_merger.*)๊ณผ ๋ง์ง ์์, vLLM 0.32๊ฐ ๋น์ ์ธต์ FP8๋ก ๋ง๋ค๊ณ BF16 ๋น์ ๊ฐ์ค์น๋ฅผ ๊ทธ๋๋ก ๋ฃ์์ โ ์ด๋ฏธ์ง ์ ๋ ฅ ๋ต์ด ํ๋ฆผ(ํ ์คํธ ์ ๋ ฅ์ ์ํฅ ์์) โconfig.json๋ง ์์ , ๋ฌด์ ๋ชฉ๋ก์ vLLM ์ด๋ฆ ๊ท์น(re:vpm\..*,re:vit_merger\..*)์ ์ถ๊ฐํ๊ณ ๊ธฐ์กด ๊ท์นre:model\.visual\..*ยทre:model\.vision_tower\..*ยทre:model\.merger\..*๋ฅผre:(model\.)?โฆํํ๋ก ๋ํ(๊ฐ์ค์น ๋์ผ) โ vLLM 0.32.0 ํ์ธ(2026-11-02): ๋น์ ๊ฐ์ค์น๊ฐ BF16์ผ๋ก ๋ก๋๋จ, ๋ชจ๋ธ์ด ์ํ ๊ทธ๋ฆผ(๋นจ๊ฐ ์ยทํ๋ ์ฌ๊ฐํยท๊ธ์ "QWEN TEST 2026")์ ๋ฐ๋ฅด๊ฒ ์ค๋ช (vLLM 0.32.0 + transformers 5.21์์๋ ์ด๋ฏธ์ง ์ ์ฒ๋ฆฌ๊ฐ ์คํจํ๋ฏ๋ก โ ์๊ตฌ์ฌํญ ์ฐธ๊ณ โ ์ด ์ ๊ฒ๋ง ์ด๋ฏธ์ง ์ฒ๋ฆฌ๊ธฐ์ ๋น ์งdownsample_mode์ธ์๋ฅผ ๊ธฐ๋ณธ๊ฐ์ผ๋ก ์ฑ์ ์คํ) โ 2026-11-01 23:41:44 UTC(๋ฆฌ๋น์ a1f77bb) ์ด์ ์ ๋ฐ์๋ค๋ฉดconfig.json๋ง ๋ค์ ๋ฐ์ ๊ฒ
๋ผ์ด์ ์ค
- ๋ผ์ด์ ์ค: Apache-2.0 โ ์๋ณธ ๋ชจ๋ธ์ ๋ผ์ด์ ์ค๋ฅผ ๊ทธ๋๋ก ๋ฐ๋ฆ
- ๊ถ๋ฆฌ: ์๋ณธ ๊ฐ์ค์นยทํ ํฌ๋์ด์ ๋ OpenBMB ์์ , ์ด ๋ฆฌํฌ๋ ์์ํ ํ์๋ฌผ
- Downloads last month
- 196
Model tree for axteam/MiniCPM-V-4.7-35B-A3B-FP8-dynamic
Base model
openbmb/MiniCPM-V-4.7-35B-A3B