LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

BgGPT 7B Instruct V0.2 by INSAIT-Institute

 »  All LLMs  »  INSAIT-Institute  »  BgGPT 7B Instruct V0.2   URL Share it on

  43010   
  Autotrain compatible Base model:finetune:mistralai/... Base model:mistralai/mistral-7...   Bg   Bggpt   Conversational   En   Endpoints compatible   Insait   Instruct   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/INSAIT-Institute/BgGPT-7B-Instruct-v0.2 

BgGPT 7B Instruct V0.2 Benchmarks

ARC: 60.58 vs 96.7 (so35)-37.4%
HellaSwag: 82.18 vs 95.3 (gpt4)-13.8%
MMLU: 60.5 vs 88.3 (so35)-31.5%
TruthfulQA: 54.63 vs 59 (gpt4)-7.4%
WinoGrande: 76.48 vs 87.5 (gpt4)-12.6%
GSM8K: 44.12 vs 96.4 (so35)-54.2%
LLME Score: 0.21821
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
BgGPT 7B Instruct V0.2 (INSAIT-Institute/BgGPT-7B-Instruct-v0.2)
🌟 Advertise your project 🚀

BgGPT 7B Instruct V0.2 Parameters and Internals

Model Type 
Causal decoder-only transformer language model
Additional Notes 
The model is continuously pretrained to gain Bulgarian language and cultural understanding, leveraging multiple datasets including Bulgarian web crawl data and translations of English datasets. The tokenizer is optimized for Bulgarian Cyrillic text.
Supported Languages 
bg (High proficiency), en (Retained proficiency)
Input Output 
Input Format:
[INST] and [/INST] tokens around prompts
LLM NameBgGPT 7B Instruct V0.2
Repository 🤗https://huggingface.co/INSAIT-Institute/BgGPT-7B-Instruct-v0.2 
Base Model(s)  mistralai/Mistral-7B-v0.1   mistralai/Mistral-7B-v0.1
Model Size7b
Required VRAM14.6 GB
Updated2025-09-20
MaintainerINSAIT-Institute
Model Typemistral
Instruction-BasedYes
Model Files  10.0 GB: 1-of-2   4.6 GB: 2-of-2
Supported Languagesbg en
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size38000
Torch Data Typebfloat16

Best Alternatives to BgGPT 7B Instruct V0.2

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB14418
SpydazWeb AI HumanAI RP512K / 14.4 GB151
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
... Summarize 64K QLoRANET Merged128K / 4.1 GB60
...1 Summarize 64K LoRANET Merged128K / 14.4 GB100
Mistral 7B Instruct V0.232K / 14.4 GB33160952630
Mistral 7B Instruct V0.132K / 14.4 GB1523071797
Seed X Instruct 7B32K / 15 GB1715123
Mixtral AI CyberCoder32K / 14.3 GB00
Note: green Score (e.g. "73.2") means that the model is better than INSAIT-Institute/BgGPT-7B-Instruct-v0.2.

Rank the BgGPT 7B Instruct V0.2 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 51483 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124
  Support LLM Explorer