Sign in

Is DeepSeek V4 Pro Preview SOTA in Factuality?#69 in FactualityRanked #69 in Factuality.

How it scores here

DeepSeek V4 Pro Preview in Factuality

Primary benchmarks

  • LMArena Text factuality: 1448 (CI 1444–1451) on Nov 19, 2026 (#79 of 215).

On the category's weighted score it is 40.6 points behind #1 Gemini 4 Carbon and 0.6 behind the next model up, Claude Sonnet 4.5.

Nearby on the board

#ModelCategory score
66Gemma 4 31B61.2
67Step 5 Preview60.6
68Claude Sonnet 4.560.0
69DeepSeek V4 Pro Preview59.4
70Hy358.8
71GLM 4.658.2
72DeepSeek V3.2 Exp57.6
Standings

Factuality

#ModelState
1Gemini 4 CarbonCo-SOTA
1Gemini 4 NeonCo-SOTA
2Claude Opus 5.5Near-SOTA
3Claude Fable 5.1Near-SOTA
4Muse WonderNear-SOTA
5Gemini 4 HydrogenNear-SOTA
6Claude Opus 4.6Near-SOTA
7Gemini 4 ArgonWas SOTA
8Claude Opus 5Was SOTA
9Mimo V2.6 Pro#9 in this category
10GPT-5.5Was SOTA
11GPT-5.4Was SOTA

Category page →