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

T Llama by 1TuanPham

 »  All LLMs  »  1TuanPham  »  T Llama   URL Share it on

  843   

T Llama is an open-source language model by 1TuanPham. Features: 6.8b LLM, VRAM: 13.8GB, Context: 2K, License: llama2, LLM Explorer Score: 0.16, Arc: 54.2, HellaSwag: 76.5, MMLU: 48, GSM8K: 29.6.

  En   Endpoints compatible   Llama   Model-index   Pytorch   Region:us   Safetensors   Sharded   Tensorflow   Vi
Model Card on HF 🤗: https://huggingface.co/capleaf/T-Llama 

T Llama Benchmarks

ARC: 54.18 vs 96.7 (so35)-44%
HellaSwag: 76.48 vs 95.3 (gpt4)-19.7%
MMLU: 47.98 vs 88.3 (so35)-45.7%
GSM8K: 29.64 vs 96.4 (so35)-69.3%
LLME Score: 0.15627
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
T Llama (capleaf/T-Llama)
🌟 Advertise your project 🚀

T Llama Parameters and Internals

Model Type 
Llama2-7B Decoder-only
Additional Notes 
The model is a proof of effort that a single person can achieve state-of-the-art results by fine-tuning their own model.
Supported Languages 
English (full), Vietnamese (full)
Training Details 
Data Sources:
BactrianX, OpenOrca_translated, WizardLM_70k_translated, TigerLabMathInstruct_translated_vi, GradeSchoolMathInstruct_translated, vilm_lima-vi, MTEngVietnamese, databricks_dolly15k_translated, AlpacaCleaned_translated, databricks_dolly15k, OpenOrca, GradeSchoolMathInstruct, AlpacaCleaned, WebglmQA
Data Volume:
120GB (Vietnamese data)
Methodology:
Fine-tuning with bilingual support
Training Time:
~47.5 days Approx*
Hardware Used:
GPU: VGA NVIDIA Tesla P100 16GB, SYSTEM RAM: 32GB
Model Architecture:
Decoder-only architecture based on Llama2-7b
LLM NameT Llama
Repository 🤗https://huggingface.co/capleaf/T-Llama 
Model Size6.8b
Required VRAM13.8 GB
Updated2026-05-19
Maintainer1TuanPham
Model Typellama
Model Files  0.9 GB: 1-of-16   0.8 GB: 2-of-16   0.9 GB: 3-of-16   0.9 GB: 4-of-16   0.9 GB: 5-of-16   0.9 GB: 6-of-16   0.9 GB: 7-of-16   0.9 GB: 8-of-16   0.9 GB: 9-of-16   0.9 GB: 10-of-16   0.9 GB: 11-of-16   0.9 GB: 12-of-16   0.9 GB: 13-of-16   0.9 GB: 14-of-16   0.9 GB: 15-of-16   0.4 GB: 16-of-16   0.9 GB: 1-of-16   0.8 GB: 2-of-16   0.9 GB: 3-of-16   0.9 GB: 4-of-16   0.9 GB: 5-of-16   0.9 GB: 6-of-16   0.9 GB: 7-of-16   0.9 GB: 8-of-16   0.9 GB: 9-of-16   0.9 GB: 10-of-16   0.9 GB: 11-of-16   0.9 GB: 12-of-16   0.9 GB: 13-of-16   0.9 GB: 14-of-16   0.9 GB: 15-of-16   0.4 GB: 16-of-16
Supported Languagesvi en
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length2048
Model Max Length2048
Transformers Version4.33.1
Tokenizer ClassLlamaTokenizer
Padding Token[PAD]
Vocabulary Size45452
Torch Data Typebfloat16

Best Alternatives to T Llama

Best Alternatives
Context / RAM
Downloads
Likes
PersianMind V1.02K / 13.7 GB113970
Note: green Score (e.g. "73.2") means that the model is better than capleaf/T-Llama.

Rank the T Llama 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? 53972 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