Vikhrmodels / text-generation updated 1 year ago

Vikhr-Nemo-12B-Instruct-R-21-09-24

Name: Vikhr-Nemo-12B-Instruct-R-21-09-24
Author: Vikhrmodels

Vikhr-Nemo - это наша флагманская унимодальная LLM (Large Language Model) представляющая из себя улучшенную версию mistralai/Mistral-Nemo-Instruct-2407 командой VikhrModels, адаптированную преимущественно для русского и английского языков. Для ее обучения мы использовали несколько этапов включающих в себя SFT и SMPO - нашу собственную вариацию DPO, подробнее читайте в секции "Как эта модель создавалась".

Params

12.3 B

Context

1,024,000

Downloads 30d

396 K

Likes

141

Commercial use: allowed apache-2.0 Not gated SAFETENSORS 2 languages View on Hugging Face ↗

Download history

daily snapshots · 10 days

398 K396 K

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	bf16	25.9 GB	30.8 GB	❌ Won’t fit
model.safetensors (bf16, full)	bf16 + 1000K ctx	25.9 GB	258.6 GB	❌ Won’t fit

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub

~ · curl · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/vikhr-nemo-12b-instruct-r-21-09-24
{
  "hf_id": "Vikhrmodels/Vikhr-Nemo-12B-Instruct-R-21-09-24",
  "params_b": 12.25,
  "context_length": 1024000,
  "license": { "id": "apache-2.0", "commercial": "yes" },
  "downloads_30d": 396277,
  "vram_estimates": [
    { "quant": "bf16", "gb": 30.8 }
  ],
  "updated_at": "2026-07-28T18:05:50Z"
}

est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture: MistralForCausalLM
Parameters: 12.3 B
Tensor type: BF16
Context length: 1,024,000
Vocabulary: 131,074
Layers / heads: 40 / 32
Licence: apache-2.0
First seen on the Hub: 2024-09-20
Training datasets: Vikhrmodels/GrandMaster-PRO-MAX, Vikhrmodels/Grounded-RAG-RU-v2
Added to our catalog: 2026-07-28

Compare with

Qwen3-8B 16.3 M Qwen2.5-7B-Instruct 12.1 M Qwen3.6-35B-A3B-NVFP4 11.3 M gpt-oss-20b 8.6 M Qwen3-32B 8.3 M Llama-3.1-8B-Instruct 8.0 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB