Fine-tuning GPT-OSS-20B for multilingual reasoning on Amazon SageMaker HyperPod

What happened

SFT took reasoning-language accuracy from 22% to 100%, and GRPO added measurable gains where headroom existed. The model receives a reasoning language: system-prompt instruction and must both reason and answer in that target language, overriding its strong English prior.

GPT-OSS-20B is an open-weight (a model whose trained parameters anyone may download) (published so anyone may download the trained model) reasoning model that thinks in English. Ask it to reason in French, German, Spanish, or Italian and it acknowledges the instruction, then produces its chain-of-thought in English anyway.

On a 50-example multilingual reasoning test, the base model reasoned in the correct target language only 22 percent of the time. That 22 percent was almost entirely the English examples (English 100 percent, French 0 percent, German 10 percent, Spanish 0 percent, Italian 0 percent).

For teams deploying reasoning models for non-English users, that gap matters. Artificial Intelligence Fine-tuning GPT-OSS-20B for multilingual reasoning on Amazon SageMaker HyperPod GPT-OSS-20B is an open-weight reasoning model that thinks in English.

A user who asks a question in French expects the model to reason and answer in French, not to switch to English mid-thought. Closing that gap is a fine-tuning (further training of an existing model for a narrower job) problem.

The short answer: SFT does most of the work, GRPO refines when there is headroom to refine, and matching the technique to the capability gap is what decides the outcome. The task and dataset The dataset is HuggingFaceH4/Multilingual-Thinking — a chain-of-thought reasoning dataset translated into French, Spanish, German, and Italian alongside English, roughly 949 training and 50 validation examples.