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

OpenBezoar SFT by SurgeGlobal

 »  All LLMs  »  SurgeGlobal  »  OpenBezoar SFT   URL Share it on

  1918   

OpenBezoar SFT is an open-source language model by SurgeGlobal. Features: 3b LLM, VRAM: 13.7GB, Context: 2K, License: cc-by-nc-4.0, Instruction-Based, HF Score: 41.3, LLM Explorer Score: 0.13, Arc: 40.9, HellaSwag: 71.2, MMLU: 28.5, TruthfulQA: 38.4, WinoGrande: 66.3, GSM8K: 2.5.

  Arxiv:2306.02707   Arxiv:2404.12195 Base model:finetune:openlm-res... Base model:openlm-research/ope... Dataset:surgeglobal/evol-instr...   Dataset:surgeglobal/lamini   Dataset:surgeglobal/orca   En   Endpoints compatible   Instruct   Llama   Pytorch   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/SurgeGlobal/OpenBezoar-SFT 

OpenBezoar SFT Benchmarks

ARC: 40.87 vs 96.7 (so35)-57.7%
HellaSwag: 71.24 vs 95.3 (gpt4)-25.2%
MMLU: 28.46 vs 88.3 (so35)-67.8%
TruthfulQA: 38.44 vs 59 (gpt4)-34.8%
WinoGrande: 66.3 vs 87.5 (gpt4)-24.2%
GSM8K: 2.5 vs 96.4 (so35)-97.4%
LLME Score: 0.13263
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
OpenBezoar SFT (SurgeGlobal/OpenBezoar-SFT)
🌟 Advertise your project 🚀

OpenBezoar SFT Parameters and Internals

Model Type 
instruction-following, text generation
Additional Notes 
The model uses Q-LoRA with a configuration of r: 16, alpha: 16, dropout: 0.05 on target modules [q_proj, v_proj, k_proj]. Uses datasets LaMini, Orca, Evol-Instruct for instruction tuning.
Supported Languages 
en (full)
Input Output 
Input Format:
Modified Alpaca prompt template
Accepted Modalities:
text
Output Format:
Text
Performance Tips:
Use the prescribed prompt template for optimal results
LLM NameOpenBezoar SFT
Repository 🤗https://huggingface.co/SurgeGlobal/OpenBezoar-SFT 
Base Model(s)  Open Llama 3b V2   openlm-research/open_llama_3b_v2
Model Size3b
Required VRAM13.7 GB
Updated2026-04-11
MaintainerSurgeGlobal
Model Typellama
Instruction-BasedYes
Model Files  10.0 GB: 1-of-2   3.7 GB: 2-of-2   10.0 GB: 1-of-2   3.7 GB: 2-of-2
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licensecc-by-nc-4.0
Context Length2048
Model Max Length2048
Transformers Version4.33.0
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typefloat32

Best Alternatives to OpenBezoar SFT

Best Alternatives
Context / RAM
Downloads
Likes
Llama 3.2 3B Instruct128K / 6.5 GB16749181724
Llama 3.2 3B Instruct128K / 6.5 GB14269
Llama 3.2 3B RP Toxic Fuse128K / 6.4 GB42
DeepSeek R1 Distill Llama 3B128K / 6.5 GB14116
Llama Email Fraud Detector128K / 6.4 GB80
Llama 3.2 3B Instruct128K / 6.5 GB12293288
Zeitgeist 3B V1128K / 6.5 GB1055
... 3.2 3B Math Instruct RE1 ORPO128K / 6.5 GB480
ReasoningCore 3B T1 1128K / 6.5 GB51
Llama 3.2 3B ToxicKod128K / 6.4 GB42
Note: green Score (e.g. "73.2") means that the model is better than SurgeGlobal/OpenBezoar-SFT.

Rank the OpenBezoar SFT 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? 52721 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.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer