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

Kona2 Small 3.8B by tbilisi-ai-lab

 »  All LLMs  »  tbilisi-ai-lab  »  Kona2 Small 3.8B   URL Share it on

  40   

Kona2 Small 3.8B is an open-source language model by tbilisi-ai-lab. Features: 3.8b LLM, VRAM: 15.8GB, Context: 128K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.21.

Base model:finetune:microsoft/... Base model:microsoft/phi-3.5-m...   Conversational   Custom code   Efficient   En   Endpoints compatible   Georgian   Instruct   Ka   Multilingual   Phi3   Region:us   Safetensors   Sharded   Small   Tensorflow
Model Card on HF 🤗: https://huggingface.co/tbilisi-ai-lab/kona2-small-3.8B 

Kona2 Small 3.8B Benchmarks

LLME Score: 0.20901
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Kona2 Small 3.8B (tbilisi-ai-lab/kona2-small-3.8B)

Kona2 Small 3.8B Parameters and Internals

LLM NameKona2 Small 3.8B
Repository 🤗https://huggingface.co/tbilisi-ai-lab/kona2-small-3.8B 
Base Model(s)  microsoft/Phi-3.5-mini-instruct   microsoft/Phi-3.5-mini-instruct
Model Size3.8b
Required VRAM15.8 GB
Updated2026-07-21
Maintainertbilisi-ai-lab
Model Typephi3
Instruction-BasedYes
Model Files  4.9 GB: 1-of-4   5.0 GB: 2-of-4   5.0 GB: 3-of-4   0.9 GB: 4-of-4
Supported Languageska en
Model ArchitecturePhi3ForCausalLM
Licenseapache-2.0
Context Length131072
Model Max Length131072
Transformers Version4.57.1
Tokenizer ClassLlamaTokenizer
Padding Token<pad>
Vocabulary Size51997

Best Alternatives to Kona2 Small 3.8B

Best Alternatives
Context / RAM
Downloads
Likes
Phi 4 Mini Instruct128K / 7.7 GB514603791
Phi 3.5 Mini Instruct128K / 7.7 GB710372997
Phi 3 Mini 128K Instruct128K / 7.7 GB2336631700
Phi 4 Mini Instruct Heretic128K / 7.7 GB120
Phi 3 Mini 128K Instruct128K / 7.7 GB100
Phi 3.5 Mini TitanFusion 0.1128K / 7.7 GB70
MediPhi Instruct128K / 7.7 GB167169
NuExtract V1.5128K / 7.7 GB10851189
ECE EIFFEL 3Bv2128K / 7.7 GB60
MediPhi Clinical128K / 7.7 GB80215
Note: green Score (e.g. "73.2") means that the model is better than tbilisi-ai-lab/kona2-small-3.8B.

Rank the Kona2 Small 3.8B 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? 55346 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, Arena and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer