Skip to content
Computer vision

Viper-9B

9B-агент для кода, функций и двух языков

2 min read

Viper-9B — флагманская модель линейки RR-Gen1 от rrchat на базе Qwen3.5-9B: 8,95 млрд параметров и поддержка русского и английского. Заточена под автономную оркестрацию инструментов, structured function calling, генерацию кода и соревновательное программирование.

Ключевые особенности

  • Агентные сценарии: нативная поддержка структурированных XML-вызовов функций, 99.0% valid call rate и 57.0% BFCL AST.
  • Код: LiveCodeBench v6 — 33.0% против 19.0% у базы, с ростом на Easy и Medium.
  • Python: 84.7% на MBPP+ (+1.4% к базе).
  • Русский и английский: MGSM ru — 44.4% (+2.4%), билингвальные инструкции.
  • R-Tuning: меньше уверенных галлюцинаций на неизвестных темах.
  • MMLU-Redux — 74.0%, общие знания сохраняются.

Пример

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
model_id = "rrchat/Viper-9B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.bfloat16, device_map="auto")
messages = [{"role": "user", "content": "Привет! Помоги решить задачу."}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
# inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
ПодробнееБенчмарки: LiveCodeBench v6 — 33.0% против 19.0% у базы; MBPP+ — 84.7%; MGSM ru — 44.4%; BFCL AST — 57.0% при 99.0% валидных вызовов.
BFCL AST чуть ниже базы (58.0%), IFEval — 54.7% (паритет ~55%).
Обучение: файнтюн Qwen3.5-9B с R-Tuning для калиброванной «честности».
Оценки локальные: greedy non-thinking, max_tokens 1024, llama-server с jinja; могут отличаться от официальных.
Требования к VRAM в карточке не указаны; есть отдельные GGUF-веса.
Discuss Telegram 1 26

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

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.

    # rrchat/Viper-9B
    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 rrchat/Viper-9B --local-dir ~/models/Viper-9B
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to Viper-9B
  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

New posts about versions of Viper-9B (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

rrchat/Viper-9B
Author
rrchat
Category
Computer vision
Released
Posted

On NeyroLearn

Hugging Face

261

Ask AI about this model

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

Comments

Other

LINK-01

LINK-01