WHAT NEW IN X10.29 THU

AI collections from X · 22 collections · 60,636 posts

benchmarkIMAGE

Flint v1.0 ranks 7th of 14 on WikiBench, ahead of MiniMax M3.1 Flash

Recursive SI's Flint v1.0 placed 7th of 14 on WikiBench with a 4.56 average, scoring only 0.91 on tool use but still finishing ahead of MiniMax M3.1 Flash.

Wayne Lowry
@wikiwayneArchived 6 days agoOriginally posted 2026.10.22
Flint v1.0 ranks 7th of 14 on WikiBench, ahead of MiniMax M3.1 Flash (image 1 of 2)
Flint v1.0 ranks 7th of 14 on WikiBench, ahead of MiniMax M3.1 Flash (image 2 of 2)
Wayne Lowry @wikiwayneOriginally posted 2026.10.22
Open original on X

Flint v1.0 from Recursive SI just ran WikiBench: 7th of 14, average 4.56. Free on OpenRouter. It's sold as a coding agent, yet tool use scored 0.91. Where it held up: • Humanized writing 8.25 • Research 7.60 • Hallucination resistance 7.50 • Security 3.75, tied for best in class It struggled on tool use (0.91) and computation (2.00). It still finished ahead of Laguna S 2.1, MiniMax M3.1 Flash, and North Mini Code. Would you run a free model that writes well but can't call tools?

x.com/wikiwayne/status/2113222390628929676Archived 2026.10.23 12:10