Synthesis history

3 versions, newest first.

Version 3 2026-06-05 18:15 UTC · 212 items

Two substantive additions this pass. First, Microsoft's official Twitter disclosed that MAI-Thinking-1 scores 52.8% on SWE Bench Pro, described as competitive with Claude Opus 4.6 Reasoning — a specific benchmark claim …
Version 2 2026-06-04 08:13 UTC · 186 items

Two substantive additions this pass. First, Microsoft's Aion 1.0 Instruct and Aion 1.0 Plan — local Windows models for on-device reasoning without cloud dependency — were absent from the initial Build synthesis and fill…
Version 1 2026-06-03 02:20 UTC · 135 items

Microsoft held Build 2026 on June 2 in San Francisco, announcing seven in-house MAI models led by MAI-Thinking-1 — a 1-trillion-parameter mixture-of-experts reasoning model claimed to outperform Claude Sonnet 4.6[^23785…