LLM EXPLORER 60,645 MODELS INDEXED

Vikhr Nemo 12B Instruct R 21 09 24 by Vikhrmodels

By Vikhrmodels · 396277 downloads
HF Repo ↗ Copy URL

Vikhr Nemo 12B Instruct R 21 09 24 is an open-source language model by Vikhrmodels. Features: 12b LLM, VRAM: 24.5GB, Context: 1000K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.31.

  Arxiv:2405.13929 Base model:finetune:mistralai/... Base model:mistralai/mistral-n...   Conversational Dataset:vikhrmodels/grandmaste... Dataset:vikhrmodels/grounded-r...   Deploy:azure   En   Endpoints compatible   Instruct   Mistral   Region:us   Ru   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Vikhrmodels/Vikhr-Nemo-12B-Instruct-R-21-09-24 

Vikhr Nemo 12B Instruct R 21 09 24 Parameters and Internals

Model Type 
large language model, text generation, instruction following, multilingual, RAG-enabled
Use Cases 
Areas:
Research, Multilingual text generation, Reasoning, Summarization, Code generation, Dialogue management
Applications:
Multilingual applications requiring English and Russian support
Primary Use Cases:
RAG systems that dynamically search and retrieve information
Limitations:
Low safety response level
Considerations:
Use with low temperature settings
Additional Notes 
High proficiency in both Russian and English NLP tasks. Safeguards should be employed due to default low safety settings.
Supported Languages 
en (high), ru (high), others (some)
Training Details 
Data Sources:
Vikhrmodels/GrandMaster-PRO-MAX, Vikhrmodels/Grounded-RAG-RU-v2
Methodology:
SFT, SMPO, Rejection Sampling
Context Length:
128000
Input Output 
Input Format:
JSON and API format
Accepted Modalities:
text
Output Format:
JSON and Text
Performance Tips:
Use low temperature settings for best performance
LLM NameVikhr Nemo 12B Instruct R 21 09 24
Repository 🤗https://huggingface.co/Vikhrmodels/Vikhr-Nemo-12B-Instruct-R-21-09-24 
Base Model(s)  Mistral Nemo Instruct 2407   mistralai/Mistral-Nemo-Instruct-2407
Model Size12b
Required VRAM24.5 GB
Updated2026-08-05
MaintainerVikhrmodels
Model Typemistral
Instruction-BasedYes
Model Files  4.9 GB: 1-of-5   4.9 GB: 2-of-5   4.9 GB: 3-of-5   4.9 GB: 4-of-5   4.9 GB: 5-of-5
Supported Languagesen ru
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length1024000
Model Max Length1024000
Transformers Version4.44.2
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<pad>
Vocabulary Size131074
Torch Data Typebfloat16

Best Alternatives to Vikhr Nemo 12B Instruct R 21 09 24

Best Alternatives
Context / RAM
Downloads
Likes
Qliphoth 12B V1.11000K / 24.5 GB223
Qliphoth 12B V11000K / 24.5 GB262
SauerkrautLM Nemo 12B Instruct1000K / 24.5 GB147322
Magnum V4 12B1000K / 24.5 GB111654
MN Slush1000K / 24.5 GB18132
ChatWaifu V1.41000K / 24.5 GB33520
...g M Claude Opus High Reasoning1000K / 24.5 GB863
ChatWaifu 12B V2.01000K / 24.5 GB17522
Mistral Nemo Bophades 12B1000K / 24.5 GB189
Mistral Nemo Wissenschaft 12B1000K / 24.5 GB228