Microsoft Launches First MAI Models; MAI-Thinking-1 Claims Claude Opus 4.6-Level Performance on SWE-Bench Pro
emmanuelvivier · x · 2026-09-16
This recap reports Microsoft releasing its first in-house MAI models, with MAI-Thinking-1 claiming parity with Claude Opus 4.6 on SWE-Bench Pro. A landmark step in Microsoft's self-developed model efforts, though the benchmark figures are self-reported and unverified.
More from Models
- New model Jev plays Super Mario Bros in real time on fast inference — hardimanjames · 2026-09-16
- Rumors: OpenAI sitting on proofs of multiple Millennium Problems over backlash fears — haider1 · 2026-09-16
- NeoHorse-1 4B beats its Qwen3.5 base on all ten tests, average score up 58.94 to 64.87 — PrajwalTomar_ · 2026-09-16
- StepFun's StepAudio 3 Realtime reasons while speaking, hits 90.6 on MMSU — stepfun-ai · 2026-09-16
- DeepSeek harness spotted with experimental computer use and browser use updates — kevinkern · 2026-09-16
- Japan's crowdsourced Minna de Honkoku OCR upgrades to v19 with better kanbun recognition — tkasasagi · 2026-09-16