Skip to content
Computer vision

LingeeJevMulti-4B-v1

мультимодальные решения вероятностями

1 min readupdated

LingeeJevMulti-4B-v1 — мультимодальная модель на 4,54 млрд параметров на базе Qwen3.5-4B. Она принимает текст или структурированное состояние и одно либо несколько изображений, а возвращает вероятности для типизированных решений choice, noul и score без генерации пояснительного текста. Лицензия — Apache-2.0.

Ключевые особенности

  • 4,54 млрд параметров, включая 0,33 млрд vision encoder.
  • На входе текст или структурированное состояние плюс одно или несколько изображений; лимит — 1 638 400 пикселей на изображение.
  • Возвращает нативные вероятности для решений choice, noul и score, без пояснительного текста.
  • HTTP API: POST /v1/systemone.
  • Для воспроизводимости: requirements.txt, Decision Readout, runtime-файлы и COMPLETE.json с SHA-256.

Пример

python systemone.py --model . --host 127.0.0.1 --port 8000
ПодробнееБенчмарки: в карточке указаны локальные замеры на H200 при 1.6MP в benchmark_1p6mp.json; официальные результаты лидерборда появятся после оценки мейнтейнерами.
Ограничение: изображения не больше 1,6 мегапикселя.
Лицензия Apache-2.0; при распространении нужно сохранить LICENSE и NOTICE.
Модель не генерирует объяснения — только распределения вероятностей.
Discuss Telegram 0 40

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 8.5 GB · ready files9.8 GB
  • FP8 / INT8recommendedweights 4.2 GB5.2 GB
  • 4-bitweights 2.4 GB3.1 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 2060 · BF16 — from RTX 3080

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # Ashan0525/LingeeJevMulti-4B-v1
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download Ashan0525/LingeeJevMulti-4B-v1 --local-dir ~/models/LingeeJevMulti-4B-v1
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to LingeeJevMulti-4B-v1
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

  1. Qwen3.5-4B-Basefamily root
  2. Qwen3.5-4B
  3. LingeeJevMulti-4B-v1

New posts about versions of Qwen3.5-4B-Base (quants, finetunes, LoRAs) will show up under the bell.

Quantised versions 12All on Hugging Face
Fine-tunes 12All on Hugging Face
LoRAs and adapters 12All on Hugging Face
Merges 1All on Hugging Face

About the model

Ashan0525/LingeeJevMulti-4B-v1
Author
Ashan0525
Category
Computer vision
Released
Posted

On NeyroLearn

Hugging Face

400

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments