0 vs 1 benchmarks won
Meta Muse Spark | Mistral Mistral Medium 3.5 | |
|---|---|---|
| Overview | ||
| Company | Meta | Mistral |
| Release date | Apr 8 2026 | Apr 29 2026 |
| Model type | — | — |
| Open source | No | Yes |
| Specifications | ||
Parameters | — | 128B |
Context window | — | 256k |
| 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. | 77.4% | 77.6%Best |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 89.5% | — |
Multimodal MMMUTests the AI on understanding images and text together across many college subjects. Higher is better. | 80.4% | — |
| Timeline | ||
| Release gap | Muse Spark shipped 21 days before Mistral Medium 3.5 | |
Mistral Medium 3.5 leads Muse Spark on 1 of the 1 benchmark they both report (SWE-Bench Verified). Muse Spark shipped 21 days before Mistral Medium 3.5, so benchmark comparisons should account for the intervening progress.
Mistral Medium 3.5 is an open-source / open-weight model; Muse Spark is proprietary.
On SWE-Bench Verified, Mistral Medium 3.5 leads at 77.6% vs Muse Spark at 77.4%.