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

Openthaigpt 1.0.0 70B Chat by openthaigpt

 »  All LLMs  »  openthaigpt  »  Openthaigpt 1.0.0 70B Chat   URL Share it on

  498   
  Arxiv:2411.07238   Autotrain compatible   En   Endpoints compatible   Finetuned   Gguf   Llama   Openthaigpt   Quantized   Region:us   Safetensors   Sharded   Tensorflow   Th
Model Card on HF 🤗: https://huggingface.co/openthaigpt/openthaigpt-1.0.0-70b-chat 

Openthaigpt 1.0.0 70B Chat Benchmarks

ARC: 64.42 vs 96.7 (so35)-33.4%
HellaSwag: 85.08 vs 95.3 (gpt4)-10.7%
MMLU: 65.53 vs 88.3 (so35)-25.8%
TruthfulQA: 54.23 vs 59 (gpt4)-8.1%
WinoGrande: 78.37 vs 87.5 (gpt4)-10.4%
GSM8K: 40.33 vs 96.4 (so35)-58.2%
LLME Score: 0.22058
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Openthaigpt 1.0.0 70B Chat (openthaigpt/openthaigpt-1.0.0-70b-chat)

Openthaigpt 1.0.0 70B Chat Parameters and Internals

Model Type 
text generation
Use Cases 
Areas:
research, commercial applications
Applications:
Thai language understanding and generation
Primary Use Cases:
chatbot and conversational AI, information retrieval augmentation (RAG)
Supported Languages 
th (Thai), en (English)
Training Details 
Data Sources:
65 billion Thai language words, 1 million Thai instruction examples
Data Volume:
65 billion Thai language words
Methodology:
Fine-tuning with 10,000 frequently used Thai words
Context Length:
4096
Input Output 
Input Format:
Llama2 Prompt Format
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Use frequent Thai words to increase generation speed.
Release Notes 
Version:
1.0.0
Date:
2024-04-08
Notes:
First release of OpenThaiGPT 70b, a 70-billion-parameter Thai language model fine-tuned for Thai instructions.
LLM NameOpenthaigpt 1.0.0 70B Chat
Repository 🤗https://huggingface.co/openthaigpt/openthaigpt-1.0.0-70b-chat 
Model Size70b
Required VRAM138.4 GB
Updated2024-12-22
Maintaineropenthaigpt
Model Typellama
Model Files  41.5 GB   9.5 GB: 1-of-15   10.0 GB: 2-of-15   9.8 GB: 3-of-15   9.8 GB: 4-of-15   9.8 GB: 5-of-15   10.0 GB: 6-of-15   9.8 GB: 7-of-15   9.8 GB: 8-of-15   9.8 GB: 9-of-15   10.0 GB: 10-of-15   9.8 GB: 11-of-15   9.8 GB: 12-of-15   9.8 GB: 13-of-15   10.0 GB: 14-of-15   0.7 GB: 15-of-15
Supported Languagesth en
GGUF QuantizationYes
Quantization Typegguf
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length2048
Model Max Length2048
Transformers Version4.35.0
Tokenizer ClassLlamaTokenizer
Padding Token<PAD>
Vocabulary Size41088
Torch Data Typebfloat16

Best Alternatives to Openthaigpt 1.0.0 70B Chat

Best Alternatives
Context / RAM
Downloads
Likes
Reflection Llama 3.1 70B Bf16128K / 141.9 GB2836
Reflection Llama 3.1 70B GGUF128K / 26.4 GB1494
...Horizon AI Korean Advanced 70B128K / 141.9 GB510
Midnight Miqu 70B V1.0 GGUF31K / 29.9 GB2674
...qu 1 70B 24GB VRAM IQ2 XS SOTA31K / 20.3 GB100
...ma3 70B Chinese Chat GGUF 4bit8K / 40 GB93218
Llama 3 70B Quantised8K / 48.7 GB122
...3 Mega Dolphin 2.9.1 120b GGUF8K / 18.4 GB51
Meta Llama 3 70B Instruct8K / 40.3 GB101
CodeLlama 70B Instruct Hf GGUF4K / 25.5 GB1742
Note: green Score (e.g. "73.2") means that the model is better than openthaigpt/openthaigpt-1.0.0-70b-chat.

Rank the Openthaigpt 1.0.0 70B Chat 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? 40066 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 v20241217
  Support LLM Explorer