The smallest OpenThai model — Thai chat on an ordinary laptop (~8 GB RAM). Fine-tuned on 2M+ Thai instruction pairs; multi-turn, RAG and tool calling.

ollama run openthai/openthai-1.5-7b

curl http://localhost:11434/api/chat \
  -d '{
    "model": "openthai/openthai-1.5-7b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='openthai/openthai-1.5-7b',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'openthai/openthai-1.5-7b',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 8 hours ago

8 hours ago

38e3a01982fb · 4.7GB ·

model

archqwen2

parameters7.62B

quantizationQ4_K_M

4.7GB

system

You are OpenThai, a Thai language assistant developed by the OpenThai project (AIEAT and iApp Techno

225B

license

See https://huggingface.co/openthaigpt — licence metadata is other/qwen. OpenThai (formerly OpenTh

199B

params

{ "num_ctx": 32768, "stop": [ "<|im_start|>", "<|im_end|>" ], "tempe

105B

template

{{ if .System }}<|im_start|>system {{ .System }}<|im_end|> {{ end }}{{ if .Prompt }}<|im_start|>user

182B

Readme

OpenThai 1.5 7B 🇹🇭

Thai chat on an ordinary laptop — the easiest place to start.

ollama run openthai/openthai-1.5-7b

The smallest OpenThai model: fine-tuned on 2,000,000+ Thai instruction pairs, with multi-turn conversation, RAG compatibility and tool calling. Runs in ~8 GB of RAM — no GPU required.

Which OpenThai model do I want?

Need	Model
🧠 Reasoning, maths, code	openthai-r1-32b
💬 Best general quality	openthai-1.6-72b
⚖️ Thai law with citations	openthai-2.0-legal
💻 Runs anywhere	this one

Tips

At 7B it trades accuracy for reach — for anything where correctness matters more than convenience, use a larger model
OpenAI-compatible API at http://localhost:11434/v1
Pair with OpenThaiRAG for retrieval

Links

Website · GGUF weights (Q4/Q5/Q8) · Paper (arXiv:2411.07238) · Discord

Free and open-weight, from the OpenThai project (formerly OpenThaiGPT) — AIEAT × iApp Technology.

# OpenThai 1.5 7B 🇹🇭

**Thai chat on an ordinary laptop — the easiest place to start.**

```
ollama run openthai/openthai-1.5-7b
```

The smallest OpenThai model: fine-tuned on **2,000,000+ Thai instruction pairs**, with
multi-turn conversation, RAG compatibility and tool calling. Runs in **~8 GB of RAM** —
no GPU required.

## Which OpenThai model do I want?

| Need | Model |
|---|---|
| 🧠 Reasoning, maths, code | [openthai-r1-32b](https://ollama.com/openthai/openthai-r1-32b) |
| 💬 Best general quality | [openthai-1.6-72b](https://ollama.com/openthai/openthai-1.6-72b) |
| ⚖️ Thai law with citations | [openthai-2.0-legal](https://ollama.com/openthai/openthai-2.0-legal) |
| 💻 Runs anywhere | **this one** |

## Tips

- At 7B it trades accuracy for reach — for anything where correctness matters more than
  convenience, use a larger model
- OpenAI-compatible API at `http://localhost:11434/v1`
- Pair with [OpenThaiRAG](https://github.com/OpenThaiGPT/openthairag) for retrieval

## Links

[Website](https://openthai.aieat.or.th/ollama) ·
[GGUF weights (Q4/Q5/Q8)](https://huggingface.co/openthaigpt/openthaigpt1.5-7b-instruct-GGUF) ·
[Paper (arXiv:2411.07238)](https://arxiv.org/abs/2411.07238) ·
[Discord](https://discord.gg/7KDdKkBGUs)

*Free and open-weight, from the OpenThai project (formerly OpenThaiGPT) — AIEAT × iApp Technology.*

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)