# Best open-weights AI model for factuality — October 2026

As of Oct 14, 2026, Mimo V2.6 Pro is state of the art in Factuality, open weights (since Sep 25, 2026; sources LMArena Text factuality; methodology v1.0).

Leading, crown pending confirmation: Kimi K3.

Category board: https://isitsota.com/c/factuality?slice=open-weights

## Top 10

- #1 Mimo V2.6 Pro
- #2 DeepSeek V4.1 Flash
- #3 GLM 5.3
- #1 Kimi K3
- #4 Mimo V2.5 Pro
- #5 GLM 5.3 Flash
- #6 GLM 5.2
- #7 GLM 5.1
- #8 Kimi K2 6
- #9 DeepSeek V4 Pro

## Crown changes

- Sep 25, 2026 — Factuality: Mimo V2.6 Pro takes the crown

## Weights

- SimpleQA Verified: primary, weight 1
- LMArena Text factuality: primary, weight 0.8