Mistral Medium 3.5vsgpt-oss-20b
Mistral Medium 3.5 | gpt-oss-20b | |
|---|---|---|
| Specifications | ||
Parameters | 128B | 21B |
Context window | 256k | 128k |
| Benchmarks | ||
Coding SWE-Bench VerifiedReal coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 77.6%Best | 60.7% |
Multidisciplinary reasoning Humanity's Last Exam · no toolsHumanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better. | — | 10.9% |
Multidisciplinary reasoning Humanity's Last Exam · with toolsHumanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | — | 17.3% |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 71.5% |
General knowledge MMLUA 57-subject multiple-choice exam — history, law, medicine, maths — that was the standard measure of how much a model knows from 2020 until roughly 2024, when frontier scores crowded into the high 80s and labs moved on to harder tests. The scores here were published years apart under different testing setups, so read them as a historical record rather than a like-for-like ranking. Higher is better. | — | 85.3% |
Community preference (code) Arena Elo (Code)Like the text arena, but people vote on which AI writes better code. The votes become a chess-style Elo rating on arena.ai. Higher is better. | 1267 | — |
| Overview | ||
| Company | Mistral | OpenAI |
| Release date | Apr 29 2026 | Aug 5 2025 |
| Access | Open Weight | Open Weight |
Which is better: Mistral Medium 3.5 or gpt-oss-20b?
Mistral Medium 3.5 leads gpt-oss-20b on 1 of the 1 benchmark they both report (SWE-Bench Verified). gpt-oss-20b shipped 267 days before Mistral Medium 3.5, so benchmark comparisons should account for the intervening progress.
Mistral Medium 3.5 has 128B parameters, while gpt-oss-20b has 21B. Context windows are 256k (Mistral Medium 3.5) vs 128k (gpt-oss-20b).
On SWE-Bench Verified, Mistral Medium 3.5 leads at 77.6% vs gpt-oss-20b at 60.7%.
Frequently asked questions
Mistral Medium 3.5 was released by Mistral on Apr 29 2026.
gpt-oss-20b was released by OpenAI on Aug 5 2025.
Mistral Medium 3.5 leads on SWE-Bench Verified — Mistral Medium 3.5 77.6% vs gpt-oss-20b 60.7%.
Mistral Medium 3.5 has a 256k context window; gpt-oss-20b has 128k.
Other comparisons
Mistral Medium 3.5vsClaude Opus 5gpt-oss-20bvsClaude Opus 5Mistral Medium 3.5vsGemini 3.6 Flashgpt-oss-20bvsGemini 3.6 FlashMistral Medium 3.5vsMuse Glimmergpt-oss-20bvsMuse GlimmerMistral Medium 3.5vsGrok 4.5gpt-oss-20bvsGrok 4.5Mistral Medium 3.5vsDeepSeek-V4-Flash-0731gpt-oss-20bvsDeepSeek-V4-Flash-0731Mistral Medium 3.5vsKimi K3gpt-oss-20bvsKimi K3