GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
tcclaviger
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
tcclaviger
Model Tree
tokyotech-llm/Medical-GPT-OSS-Swallow-120B
Input Modalities
Output Modalities
Supported Functionality
tokyotech-llm/GPT-OSS-Swallow-120B-RL-v0.1The following results report this medical-domain model on medical benchmarks. General benchmark results are intentionally omitted because this release focuses on medical-domain performance.
| Model | IgakuQA | JJSIMQA | JMMLU Medical | MMLU_Medical_JP | MedMCQA_JP | MedQA_JP | JUSMLEQA_JP | YakugakuQA |
|---|---|---|---|---|---|---|---|---|
Medical-GPT-OSS-Swallow-120B | 0.7048 | 0.6659 | 0.7022 | 0.7320 | 0.5230 | 0.5118 | 0.5584 | 0.6031 |
This model is expected to work with Hugging Face Transformers and vLLM-compatible inference stacks.
sh
vllm serve tokyotech-llm/Medical-GPT-OSS-Swallow-120B \ --tensor-parallel-size 8 \ --max-model-len 32768Once the server is running, you can send requests using an OpenAI-compatible client.
python
from openai import OpenAI model_name = "tokyotech-llm/Medical-GPT-OSS-Swallow-120B"client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY") result = client.chat.completions.create( model=model_name, messages=[ {"role": "user", "content": "日本語で、臨床現場における生成AI利用時の注意点を説明してください。"} ], max_tokens=2048, temperature=0.6, top_p=0.95, extra_body={ "top_k": 20, "min_p": 0, },) print(result.choices[0].message.content)We recommend using the generation parameters specified in generation_config.json when available. For GPT-OSS-Swallow models, commonly used settings include temperature=0.6, top_p=0.95, top_k=20, and min_p=0.
We also recommend specifying a maximum context length of 32,768 tokens or less for inference unless your serving stack has been validated with a longer context.
For large-scale inference, vLLM is recommended. Adjust --tensor-parallel-size, --gpu-memory-utilization, and --max-model-len according to the available GPU memory.
This model was adapted from GPT-OSS-Swallow-120B-RL-v0.1 using a mixture that emphasizes medical-domain text while retaining general-domain data. The medical-domain data includes resources such as biomedical literature, medical synthetic data, medical QA-style data, and clinical guideline-style text.
This model is intended for research and development. It has not been validated as a medical device and must not be used as a substitute for professional medical judgment. Outputs may contain factual errors, unsafe recommendations, or unsupported clinical claims. Any clinical use requires careful human review, validation, and compliance with applicable laws, regulations, and institutional policies.
Apache License 2.0
If you find our work helpful, please feel free to cite these papers. The Qwen3-Swallow and GPT-OSS-Swallow Technical Paper (Training Details) will be released in March.
bibtex
@inproceedings{ fujii2024continual, title={Continual Pre-Training for Cross-Lingual {LLM} Adaptation: Enhancing Japanese Language Capabilities}, author={Kazuki Fujii and Taishi Nakamura and Mengsay Loem and Hiroki Iida and Masanari Ohi and Kakeru Hattori and Hirai Shota and Sakae Mizuki and Rio Yokota and Naoaki Okazaki}, booktitle={First Conference on Language Modeling}, year={2024}}bibtex
@inproceedings{ ma2025building, title={Building Instruction-Tuning Datasets from Human-Written Instructions with Open-Weight Large Language Models}, author={Youmi Ma and Sakae Mizuki and Kazuki Fujii and Taishi Nakamura and Masanari Ohi and Hinari Shimada and Taihei Shiotani and Koshiro Saito and Koki Maeda and Kakeru Hattori and Takumi Okamoto and Shigeki Ishida and Rio Yokota and Hiroya Takamura and Naoaki Okazaki}, booktitle={Second Conference on Language Modeling}, year={2025}}[OpenAI, 2025] OpenAI. gpt-oss-120b & gpt-oss-20b Model Card, arXiv:2508.10925.
This work builds on GPT-OSS and GPT-OSS-Swallow. We thank the OpenAI team and the contributors to the GPT-OSS-Swallow project.
この成果は、国立研究開発法人新エネルギー・産業技術総合開発機構（ＮＥＤＯ）の助成事業（JPNP25006）の結果得られたものです。
This model is based on the results obtained from the project, JPNP25006, subsidized by the New Energy and Industrial Technology Development Organization (NEDO).
vllm serve tokyotech-llm/Medical-GPT-OSS-Swallow-120B \ --tensor-parallel-size 8 \ --max-model-len 32768from openai import OpenAI model_name = "tokyotech-llm/Medical-GPT-OSS-Swallow-120B"client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY") result = client.chat.completions.create( model=model_name, messages=[ {"role": "user", "content": "日本語で、臨床現場における生成AI利用時の注意点を説明してください。"} ], max_tokens=2048, temperature=0.6, top_p=0.95, extra_body={ "top_k": 20, "min_p": 0, },) print(result.choices[0].message.content)@inproceedings{ fujii2024continual, title={Continual Pre-Training for Cross-Lingual {LLM} Adaptation: Enhancing Japanese Language Capabilities}, author={Kazuki Fujii and Taishi Nakamura and Mengsay Loem and Hiroki Iida and Masanari Ohi and Kakeru Hattori and Hirai Shota and Sakae Mizuki and Rio Yokota and Naoaki Okazaki}, booktitle={First Conference on Language Modeling}, year={2024}}@inproceedings{ ma2025building, title={Building Instruction-Tuning Datasets from Human-Written Instructions with Open-Weight Large Language Models}, author={Youmi Ma and Sakae Mizuki and Kazuki Fujii and Taishi Nakamura and Masanari Ohi and Hinari Shimada and Taihei Shiotani and Koshiro Saito and Koki Maeda and Kakeru Hattori and Takumi Okamoto and Shigeki Ishida and Rio Yokota and Hiroya Takamura and Naoaki Okazaki}, booktitle={Second Conference on Language Modeling}, year={2025}}