# 🇹🇭 OpenThaiGPT 72b 1.5 Instruct ![OpenThaiGPT](https://1173516064-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FvvbWvIIe82Iv1yHaDBC5%2Fuploads%2Fb8eiMDaqiEQL6ahbAY0h%2Fimage.png?alt=media&token=6fce78fd-2cca-4c0a-9648-bd5518e644ce) [More Info](https://openthaigpt.aieat.or.th/) 🇹🇭 **OpenThaiGPT 72b Version 1.5** is an advanced 72-billion-parameter Thai language chat model based on Qwen v2.5 released on September 30, 2024. It has been specifically fine-tuned on over 2,000,000 Thai instruction pairs and is capable of answering Thai-specific domain questions.

## Online Demo: https://demo72b.aieat.or.th/ ## Example code for API Calling https://github.com/OpenThaiGPT/openthaigpt1.5_api_examples ## Highlights - **State-of-the-art Thai language LLM**, achieving the highest average scores across various Thai language exams compared to other open-source Thai LLMs. - **Multi-turn conversation support** for extended dialogues. - **Retrieval Augmented Generation (RAG) compatibility** for enhanced response generation. - **Impressive context handling**: Processes up to 131,072 tokens of input and generates up to 8,192 tokens, enabling detailed and complex interactions. - **Tool calling support**: Enables users to efficiently call various functions through intelligent responses. ## Benchmark on [OpenThaiGPT Eval](https://huggingface.co/datasets/openthaigpt/openthaigpt_eval) ** Please take a look at ``openthaigpt/openthaigpt1.5-72b-instruct`` for this model's evaluation result. | **Exam names** | **scb10x/llama-3-typhoon-v1.5x-70b-instruct** | **meta-llama/Llama-3.1-70B-Instruct** | **Qwen/Qwen2.5-72B-Instruct** | **openthaigpt/openthaigpt1.5-72b-instruct** | |:------------------------------:|:---------------------------------------------:|:-------------------------------------:|:-----------------------------:|:----------------------------------:| | **01_a_level** | 59.17% | 61.67% | 75.00% | 76.67% | | **02_tgat** | 46.00% | 40.00% | 48.00% | 46.00% | | **03_tpat1** | 52.50% | 50.00% | 55.00% | 55.00% | | **04_investment_consult** | 60.00% | 52.00% | 80.00% | 72.00% | | **05_facebook_beleble_th_200** | 87.50% | 88.00% | 90.00% | 90.00% | | **06_xcopa_th_200** | 84.50% | 85.50% | 90.00% | 90.50% | | **07_xnli2.0_th_200** | 62.50% | 63.00% | 65.50% | 70.50% | | **08_onet_m3_thai** | 76.00% | 56.00% | 76.00% | 84.00% | | **09_onet_m3_social** | 95.00% | 95.00% | 90.00% | 95.00% | | **10_onet_m3_math** | 43.75% | 25.00% | 37.50% | 37.50% | | **11_onet_m3_science** | 53.85% | 61.54% | 65.38% | 73.08% | | **12_onet_m3_english** | 93.33% | 93.33% | 96.67% | 96.67% | | **13_onet_m6_thai** | 55.38% | 60.00% | 60.00% | 56.92% | | **14_onet_m6_math** | 41.18% | 58.82% | 23.53% | 41.18% | | **15_onet_m6_social** | 67.27% | 76.36% | 63.64% | 65.45% | | **16_onet_m6_science** | 50.00% | 57.14% | 64.29% | 67.86% | | **17_onet_m6_english** | 73.08% | 82.69% | 86.54% | 90.38% | | **Micro Average** | 69.97% | 71.09% | 75.02% | 76.73% | Thai language multiple choice exams, Test on unseen test set, Zero-shot learning. Benchmark source code and exams information: https://github.com/OpenThaiGPT/openthaigpt_eval (Updated on: 30 September 2024) ## Benchmark on [scb10x/thai_exam](https://huggingface.co/datasets/scb10x/thai_exam) | Models | **Thai Exam (Acc)** | |:----------------------------------------------------------:|:-------------------:| | **api/claude-3-5-sonnet-20240620** | 69.2 | | **openthaigpt/openthaigpt1.5-72b-instruct*** | 64.07 | | **api/gpt-4o-2024-05-13** | 63.89 | | **hugging-quants/Meta-Llama-3.1-405B-Instruct-AWQ-INT4** | 63.54 | | **openthaigpt/openthaigpt1.5-14b-instruct*** | 59.65 | | **scb10x/llama-3-typhoon-v1.5x-70b-instruct** | 58.76 | | **Qwen/Qwen2-72B-Instruct** | 58.23 | | **meta-llama/Meta-Llama-3.1-70B-Instruct** | 58.23 | | **Qwen/Qwen2.5-14B-Instruct** | 57.35 | | **api/gpt-4o-mini-2024-07-18** | 54.51 | | **openthaigpt/openthaigpt1.5-7b-instruct*** | 52.04 | | **SeaLLMs/SeaLLMs-v3-7B-Chat** | 51.33 | | **openthaigpt/openthaigpt-1.0.0-70b-chat** | 50.09 | * Evaluated by OpenThaiGPT team using [scb10x/thai_exam](https://huggingface.co/datasets/scb10x/thai_exam). (Updated on: 13 October 2024) ## Licenses * Built with Qwen * Qwen License: Allow **Research** and **Commercial uses** but if your user base exceeds 100 million monthly active users, you need to negotiate a separate commercial license. Please see LICENSE file for more information.
## Sponsors

## Supports - Official website: https://openthaigpt.aieat.or.th - Facebook page: https://web.facebook.com/groups/openthaigpt - A Discord server for discussion and support [here](https://discord.gg/rUTp6dfVUF) - E-mail: kobkrit@aieat.or.th ## Prompt Format Prompt format is based on ChatML. ``` <|im_start|>system\n{sytem_prompt}<|im_end|>\n<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant\n ``` ### System prompt: ``` คุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์ ``` ### Examples #### Single Turn Conversation Example ``` <|im_start|>system\nคุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์<|im_end|>\n<|im_start|>user\nสวัสดีครับ<|im_end|>\n<|im_start|>assistant\n ``` #### Single Turn Conversation with Context (RAG) Example ``` <|im_start|>system\nคุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์<|im_end|>\n<|im_start|>user\nกรุงเทพมหานคร เป็นเมืองหลวง นครและมหานครที่มีประชากรมากที่สุดของประเทศไทย กรุงเทพมหานครมีพื้นที่ทั้งหมด 1,568.737 ตร.กม. มีประชากรตามทะเบียนราษฎรกว่า 8 ล้านคน\nกรุงเทพมหานครมีพื้นที่เท่าไร่<|im_end|>\n<|im_start|>assistant\n ``` #### Multi Turn Conversation Example ##### First turn ``` <|im_start|>system\nคุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์<|im_end|>\n<|im_start|>user\nสวัสดีครับ<|im_end|>\n<|im_start|>assistant\n ``` ##### Second turn ``` <|im_start|>system\nคุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์<|im_end|>\n<|im_start|>user\nสวัสดีครับ<|im_end|>\n<|im_start|>assistant\nสวัสดีครับ ยินดีต้อนรับครับ คุณต้องการให้ฉันช่วยอะไรครับ?<|im_end|>\n<|im_start|>user\nกรุงเทพมหานคร ชื่อเต็มยาวๆคืออะไร<|im_end|>\n<|im_start|>assistant\n ``` ##### Result ``` <|im_start|>system\nคุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์<|im_end|>\n<|im_start|>user\nสวัสดีครับ<|im_end|>\n<|im_start|>assistant\nสวัสดีครับ ยินดีต้อนรับครับ คุณต้องการให้ฉันช่วยอะไรครับ?<|im_end|>\n<|im_start|>user\nกรุงเทพมหานคร ชื่อเต็มยาวๆคืออะไร<|im_end|>\n<|im_start|>assistant\nชื่อเต็มของกรุงเทพมหานครคือ \"กรุงเทพมหานคร อมรรัตนโกสินทร์ มหินทรายุธยา มหาดิลกภพ นพรัตนราชธานีบูรีรมย์ อุดมราชนิเวศน์มหาสถาน อมรพิมานอวตารสถิต สักกะทัตติยวิษณุกรรมประสิทธิ์\" ``` ## How to use ### Free API Service (hosted by Siam.Ai and Float16.cloud) #### Siam.AI ```bash curl https://api.aieat.or.th/v1/completions \ -H "Content-Type: application/json" \ -H "Authorization: Bearer dummy" \ -d '{ "model": ".", "prompt": "<|im_start|>system\nคุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์<|im_end|>\n<|im_start|>user\nกรุงเทพมหานครคืออะไร<|im_end|>\n<|im_start|>assistant\n", "max_tokens": 512, "temperature": 0.7, "top_p": 0.8, "top_k": 40, "stop": ["<|im_end|>"] }' ``` #### Float16 ```bash curl -X POST https://api.float16.cloud/dedicate/78y8fJLuzE/v1/chat/completions \ -H "Content-Type: application/json" \ -H "Authorization: Bearer float16-AG0F8yNce5s1DiXm1ujcNrTaZquEdaikLwhZBRhyZQNeS7Dv0X" \ -d '{ "model": "openthaigpt/openthaigpt1.5-7b-instruct", "messages": [ { "role": "system", "content": "คุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์" }, { "role": "user", "content": "สวัสดี" } ] }' ``` ### OpenAI Client Library (Hosted by VLLM, please see below.) ```python import openai # Configure OpenAI client to use vLLM server openai.api_base = "http://127.0.0.1:8000/v1" openai.api_key = "dummy" # vLLM doesn't require a real API key prompt = "<|im_start|>system\nคุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์<|im_end|>\n<|im_start|>user\nกรุงเทพมหานครคืออะไร<|im_end|>\n<|im_start|>assistant\n" try: response = openai.Completion.create( model=".", # Specify the model you're using with vLLM prompt=prompt, max_tokens=512, temperature=0.7, top_p=0.8, top_k=40, stop=["<|im_end|>"] ) print("Generated Text:", response.choices[0].text) except Exception as e: print("Error:", str(e)) ``` ### Huggingface ```python from transformers import AutoModelForCausalLM, AutoTokenizer model_name = "openthaigpt/openthaigpt1.5-72b-instruct" model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype="auto", device_map="auto" ) tokenizer = AutoTokenizer.from_pretrained(model_name) prompt = "ประเทศไทยคืออะไร" messages = [ {"role": "system", "content": "คุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์"}, {"role": "user", "content": prompt} ] text = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True ) model_inputs = tokenizer([text], return_tensors="pt").to(model.device) generated_ids = model.generate( **model_inputs, max_new_tokens=512 ) generated_ids = [ output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids) ] response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] ``` ### vLLM 1. Install VLLM (https://github.com/vllm-project/vllm) 2. Run server ```bash vllm serve openthaigpt/openthaigpt1.5-72b-instruct --tensor-parallel-size 4 ``` * Note, change ``--tensor-parallel-size 4`` to the amount of available GPU cards. If you wish to enable tool calling feature, add ``--enable-auto-tool-choice --tool-call-parser hermes`` into command. e.g., ```bash vllm serve openthaigpt/openthaigpt1.5-72b-instruct --tensor-parallel-size 4 --enable-auto-tool-choice --tool-call-parser hermes ``` 3. Run inference (CURL example) ```bash curl -X POST 'http://127.0.0.1:8000/v1/completions' \ -H 'Content-Type: application/json' \ -d '{ "model": ".", "prompt": "<|im_start|>system\nคุณคือผู้ช่วยตอบคำถามที่ฉลาดและซื่อสัตย์<|im_end|>\n<|im_start|>user\nสวัสดีครับ<|im_end|>\n<|im_start|>assistant\n", "max_tokens": 512, "temperature": 0.7, "top_p": 0.8, "top_k": 40, "stop": ["<|im_end|>"] }' ``` ### Processing Long Texts The current `config.json` is set for context length up to 32,768 tokens. To handle extensive inputs exceeding 32,768 tokens, we utilize [YaRN](https://arxiv.org/abs/2309.00071), a technique for enhancing model length extrapolation, ensuring optimal performance on lengthy texts. For supported frameworks, you could add the following to `config.json` to enable YaRN: ```json { ..., "rope_scaling": { "factor": 4.0, "original_max_position_embeddings": 32768, "type": "yarn" } } ``` ### Tool Calling The Tool Calling feature in OpenThaiGPT 1.5 enables users to efficiently call various functions through intelligent responses. This includes making external API calls to retrieve real-time data, such as current temperature information, or predicting future data simply by submitting a query. For example, a user can ask OpenThaiGPT, “What is the current temperature in San Francisco?” and the AI will execute a pre-defined function to provide an immediate response without the need for additional coding. This feature also allows for broader applications with external data sources, including the ability to call APIs for services such as weather updates, stock market information, or data from within the user’s own system. #### Example: ```python import openai def get_temperature(location, date=None, unit="celsius"): """Get temperature for a location (current or specific date).""" if date: return {"temperature": 25.9, "location": location, "date": date, "unit": unit} return {"temperature": 26.1, "location": location, "unit": unit} tools = [ { "name": "get_temperature", "description": "Get temperature for a location (current or by date).", "parameters": { "location": "string", "date": "string (optional)", "unit": "enum [celsius, fahrenheit]" }, } ] messages = [{"role": "user", "content": "อุณหภูมิที่ San Francisco วันนี้ีและพรุ้่งนี้คือเท่าไร่?"}] # Simulated response flow using OpenThaiGPT Tool Calling response = openai.ChatCompletion.create( model=".", messages=messages, tools=tools, temperature=0.7, max_tokens=512 ) print(response) ``` **Full example**: https://github.com/OpenThaiGPT/openthaigpt1.5_api_examples/blob/main/api_tool_calling_powered_by_siamai.py ### GPU Memory Requirements | **Number of Parameters** | **FP 16 bits** | **8 bits (Quantized)** | **4 bits (Quantized)** | **Example Graphic Card for 4 bits** | |------------------|----------------|------------------------|------------------------|---------------------------------------------| | **7b** | 24 GB | 12 GB | 6 GB | Nvidia RTX 4060 8GB | | **13b** | 48 GB | 24 GB | 12 GB | Nvidia RTX 4070 16GB | | **72b** | 192 GB | 96 GB | 48 GB | Nvidia RTX 4090 24GB x 2 cards | ### Authors * Sumeth Yuenyong (sumeth.yue@mahidol.edu) * Kobkrit Viriyayudhakorn (kobkrit@aieat.or.th) * Apivadee Piyatumrong (apivadee.piy@nectec.or.th) * Jillaphat Jaroenkantasima (autsadang41@gmail.com) * Thaweewat Rugsujarit (thaweewr@scg.com) * Norapat Buppodom (new@norapat.com) * Koravich Sangkaew (kwankoravich@gmail.com) * Peerawat Rojratchadakorn (peerawat.roj@gmail.com) * Surapon Nonesung (nonesungsurapon@gmail.com) * Chanon Utupon (chanon.utupon@gmail.com) * Sadhis Wongprayoon (sadhis.tae@gmail.com) * Nucharee Thongthungwong (nuchhub@hotmail.com) * Chawakorn Phiantham (mondcha1507@gmail.com) * Patteera Triamamornwooth (patt.patteera@gmail.com) * Nattarika Juntarapaoraya (natt.juntara@gmail.com) * Kriangkrai Saetan (kraitan.ss21@gmail.com) * Pitikorn Khlaisamniang (pitikorn32@gmail.com) Disclaimer: Provided responses are not guaranteed.