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

CataLlama V0.2 Instruct SFT DPO Merged by catallama

 »  All LLMs  »  catallama  »  CataLlama V0.2 Instruct SFT DPO Merged   URL Share it on

  5   

CataLlama V0.2 Instruct SFT DPO Merged is an open-source language model by catallama. Features: 8b LLM, VRAM: 16.1GB, Context: 8K, License: llama3, Instruction-Based, Merged, LLM Explorer Score: 0.14.

  Merged Model Base model:catallama/catallama... Base model:finetune:catallama/...   Ca   Catalan   Conversational Dataset:catallama/catalan-dpo-... Dataset:catallama/catalan-inst...   En   Endpoints compatible   Instruct   Llama   Llama-3   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/catallama/CataLlama-v0.2-Instruct-SFT-DPO-Merged 

CataLlama V0.2 Instruct SFT DPO Merged Benchmarks

LLME Score: 0.14007
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
CataLlama V0.2 Instruct SFT DPO Merged (catallama/CataLlama-v0.2-Instruct-SFT-DPO-Merged)
🌟 Advertise your project 🚀

CataLlama V0.2 Instruct SFT DPO Merged Parameters and Internals

Model Type 
text-generation
Use Cases 
Areas:
commercial, research
Applications:
assistant-like chat, natural language generation tasks
Primary Use Cases:
Information extraction suitable for RAG, Named Entity Recognition (NER), Translation between English and Catalan, Summarization, Sentiment analysis, Chat
Limitations:
Not intended to beat benchmarks, Focus on demonstrating techniques for augmenting LLMs for new languages
Considerations:
Developers may fine-tune Llama 3 models for languages beyond English provided they comply with the Llama 3 Community License and the Acceptable Use Policy.
Additional Notes 
The model focuses on supporting Catalan and preserving rare languages.
Supported Languages 
ca (fluent), en (fluent)
Training Details 
Data Sources:
Catalan-DPO-V2, Catalan-Instruct-V2
Methodology:
Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO)
Model Architecture:
Auto-regressive language model with an optimized transformer architecture.
Input Output 
Input Format:
follows the same prompt template as Llama-3 Instruct
Accepted Modalities:
text
Output Format:
text generation
Performance Tips:
Consider using torch.bfloat16 for better performance with Transformers library.
LLM NameCataLlama V0.2 Instruct SFT DPO Merged
Repository 🤗https://huggingface.co/catallama/CataLlama-v0.2-Instruct-SFT-DPO-Merged 
Base Model(s)  catallama/CataLlama-v0.2-Instruct-SFT   catallama/CataLlama-v0.2-Instruct-DPO   catallama/CataLlama-v0.2-Instruct-SFT   catallama/CataLlama-v0.2-Instruct-DPO
Merged ModelYes
Model Size8b
Required VRAM16.1 GB
Updated2026-05-13
Maintainercatallama
Model Typellama
Instruction-BasedYes
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   4.9 GB: 3-of-4   1.2 GB: 4-of-4
Supported Languagesca en
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.42.4
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|eot_id|>
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to CataLlama V0.2 Instruct SFT DPO Merged

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB156119
UltraLong Thinking4192K / 16.1 GB103
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB11917
Cthulhu 8B V1.41048K / 16.1 GB129
...raLong 1M Instruct Abliterated1048K / 32.1 GB109
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB205957
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
Zero Llama 3.1 8B Beta61048K / 16.1 GB41
Note: green Score (e.g. "73.2") means that the model is better than catallama/CataLlama-v0.2-Instruct-SFT-DPO-Merged.

Rank the CataLlama V0.2 Instruct SFT DPO Merged 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? 53999 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