---
base_model:
- scb10x/llama3.1-typhoon2-70b-instruct
- deepseek-ai/DeepSeek-R1-Distill-Llama-70B
library_name: transformers
license: llama3.1
tags:
- mergekit
- merge
---
**Typhoon2-DeepSeek-R1-70B**: Thai reasoning large language model. (research preview)
**Typhoon2-DeepSeek-R1-70B** is a Thai ðđð reasoning large language model with 70 billion parameters, and it is based on DeepSeek R1 70B Distill and Typhoon2 70B Instruct + SFT merged.
For more details, please see our [blog](https://blog.opentyphoon.ai/introducing-typhoon2-r1-70b-enhanced-reasoning-with-deepseek-r1-abcedaa9b7ad). Demo on [opentyphoon.ai](https://playground.opentyphoon.ai/playground).
Paper: [https://arxiv.org/abs/2502.09056](https://arxiv.org/abs/2502.09056)
*To acknowledge Meta's effort in creating the foundation model and to comply with the license, we explicitly include "llama-3.1" in the model name.
## **Key Highlights**
- **Advanced Reasoning:** Comparable to DeepSeek R1 70B Distillâs reasoning capabilities.
- **Enhanced Math & Coding Performance**: Up to 6 times more accurate than Typhoon2 70B Instruct.
- **Strong Thai Language Proficiency**: Performs similarly to Typhoon2 70B Instruct.
- **Cross-Domain Reasoning**: Adapts to various domains, going beyond math and coding.
## **Model Description**
- **Model type**: A 70B instruct decoder-only model based on Llama architecture.
- **Requirement**: transformers 4.45.0 or newer.
- **Primary Language(s)**: Thai ðđð and English ðŽð§
- **License**: [Llama 3.1 Community License](https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/LICENSE)
- **Demo**: [opentyphoon.ai](https://playground.opentyphoon.ai/playground).
## **Performance**
**Reasoning Performance**
**General Instruction-Following Performance**
## **Recommend system prompt**
More controllable | less reasoning capability
```
You are a helpful assistant named Typhoon that always reasons before answering. First, provide reasoning in English, starting with "Alright!" and ending with the `` token. Then, always respond to the user in the language they use or request. Avoid unnecessary affirmations or filler phrases such as "Alright", "Okay", etc., in the response.
```
More reasoning capability | less controllable
```
You are a helpful assistant named Typhoon that always reasons before answering. First, provide reasoning in English start with Alright!, then respond to the user in the language they use or request.
```
Highest reasoning capability | least controllable
```
# No system prompt
```
## **Usage Example**
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
model_id = "scb10x/llama3.1-typhoon2-deepseek-r1-70b-preview"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
model_id,
torch_dtype=torch.bfloat16,
device_map="auto",
)
messages = [
{"role": "system", "content": 'You are a helpful assistant named Typhoon that always reasons before answering. First, provide reasoning in English, starting with "Alright!" and ending with the `` token. Then, always respond to the user in the language they use or request. Avoid unnecessary affirmations or filler phrases such as "Alright", "Okay", etc., in the response'},
{"role": "user", "content": "āļāļģāļāļ§āļāđāļāđāļĄāļāļ§āļāļāļĩāđāļāđāļāļĒāļāļĩāđāļŠāļļāļāļāļĩāđāđāļāđāļāļāļĨāļāļđāļāļāļāļ 30 āļāļķāđāļāļŠāļēāļĄāļēāļĢāļāđāļāļĩāļĒāļāđāļāđāļāđāļ§āļĒāļāļąāļ§āđāļĨāļ 0 āđāļĨāļ° 2 āđāļāđāļēāļāļąāđāļāļāļ·āļāļāļ°āđāļĢ"},
]
input_ids = tokenizer.apply_chat_template(
messages,
add_generation_prompt=True,
return_tensors="pt"
).to(model.device)
terminators = [
tokenizer.eos_token_id,
]
outputs = model.generate(
input_ids,
max_new_tokens=512,
eos_token_id=terminators,
do_sample=True,
temperature=0.7,
top_p=0.95,
)
response = outputs[0][input_ids.shape[-1]:]
# if you want only response without thinking trace. try `tokenizer.decode(response, skip_special_tokens=True).split('')[-1]` to get only response
print(tokenizer.decode(response, skip_special_tokens=True)) # Okay, .... āļāļąāļāļāļąāđāļ āļāļģāļāļ§āļāđāļāđāļĄāļāļ§āļāļāļĩāđāļāđāļāļĒāļāļĩāđāļŠāļļāļāļāļĩāđāđāļāđāļāļāļĨāļāļđāļāļāļāļ 30 āđāļĨāļ°āđāļāļĩāļĒāļāđāļāđāļāđāļ§āļĒāļāļąāļ§āđāļĨāļ 0 āđāļĨāļ° 2 āđāļāđāļēāļāļąāđāļāļāļ·āļ 2220 boxed{2220}
```
## **Inference Server Hosting Example**
```bash
pip install vllm
vllm serve scb10x/llama3.1-typhoon2-deepseek-r1-70b-preview --tensor-parallel-size 2 --gpu-memory-utilization 0.95 --max-model-len 16384 --enforce-eager
# using at least 2 80GB gpu eg A100, H100 for hosting 70b model
# to serving longer context (90k), 4 gpu is required (and you can omit --enforce-eager to improve throughput)
# see more information at https://docs.vllm.ai/
```
## **Chat template**
This model is using the DeepSeek R1 70B Distill chat template, not the Llama 3 template. Be careful.
```
{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='') %}{%- for message in messages %}{%- if message['role'] == 'system' %}{% set ns.system_prompt = message['content'] %}{%- endif %}{%- endfor %}{{bos_token}}{{ns.system_prompt}}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<ï―Userï―>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is none %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls']%}{%- if not ns.is_first %}{{'<ï―Assistantï―><ï―toolâcallsâbeginï―><ï―toolâcallâbeginï―>' + tool['type'] + '<ï―toolâsepï―>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<ï―toolâcallâendï―>'}}{%- set ns.is_first = true -%}{%- else %}{{'\\n' + '<ï―toolâcallâbeginï―>' + tool['type'] + '<ï―toolâsepï―>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<ï―toolâcallâendï―>'}}{{'<ï―toolâcallsâendï―><ï―endâofâsentenceï―>'}}{%- endif %}{%- endfor %}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is not none %}{%- if ns.is_tool %}{{'<ï―toolâoutputsâendï―>' + message['content'] + '<ï―endâofâsentenceï―>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '' in content %}{% set content = content.split('')[-1] %}{% endif %}{{'<ï―Assistantï―>' + content + '<ï―endâofâsentenceï―>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<ï―toolâoutputsâbeginï―><ï―toolâoutputâbeginï―>' + message['content'] + '<ï―toolâoutputâendï―>'}}{%- set ns.is_output_first = false %}{%- else %}{{'\\n<ï―toolâoutputâbeginï―>' + message['content'] + '<ï―toolâoutputâendï―>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<ï―toolâoutputsâendï―>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<ï―Assistantï―>\\n'}}{% endif %}
```
## **Tool use**
We don't recommend using tool use on this model.
## **Intended Uses & Limitations**
This model is an reasoning instructional model. However, itâs still undergoing development. It incorporates some level of guardrails, but it still may produce answers that are inaccurate, biased, or otherwise objectionable in response to user prompts. We recommend that developers assess these risks in the context of their use case.
## **Follow us**
**https://twitter.com/opentyphoon**
## **Support**
**https://discord.gg/CqyBscMFpg**
## **Citation**
- If you find Typhoon 2 useful for your work, please cite it using:
```
@misc{typhoon2,
title={Typhoon 2: A Family of Open Text and Multimodal Thai Large Language Models},
author={Kunat Pipatanakul and Potsawee Manakul and Natapong Nitarach and Warit Sirichotedumrong and Surapon Nonesung and Teetouch Jaknamon and Parinthapat Pengpun and Pittawat Taveekitworachai and Adisai Na-Thalang and Sittipong Sripaisarnmongkol and Krisanapong Jirayoot and Kasima Tharnpipitchai},
year={2024},
eprint={2412.13702},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2412.13702},
}
```