Sign in

Is DeepSeek V4 Flash Preview SOTA in Reasoning?#94 in ReasoningRanked #94 in Reasoning.

How it scores here

DeepSeek V4 Flash Preview in Reasoning

Primary benchmarks

  • LMArena Text (style-controlled, hard prompts): 1458 (CI 1454–1463) on Nov 19, 2026 (#118 of 263).

On the category's weighted score it is 55.9 points behind #1 Gemini 4 Carbon and 0.6 behind the next model up, Qwen3 Max Preview.

Nearby on the board

#ModelCategory score
91Longcat Flash Chat Exp45.9
92DeepSeek V4 Flash45.3
93Qwen3 Max Preview44.7
94DeepSeek V4 Flash Preview44.1
95Gemini 2.5 Pro43.5
96Mimo V2 Omni42.9
97Claude Opus 442.4
Standings

Reasoning

#ModelState
1Gemini 4 CarbonCo-SOTA
1Gemini 4 ArgonCo-SOTA
1Gemini 4 NeonCo-SOTA
2Muse WonderNear-SOTA
3Claude Opus 4.6Near-SOTA
4Claude Opus 5.5Near-SOTA
5Claude Fable 5Near-SOTA
6Claude Opus 4.7Was SOTA
7Claude Fable 5.1Was SOTA
8Muse Spark 1.3#8 in this category
9Gemini 3.8 Flash#9 in this category
10Claude Fable 5.2#10 in this category

Category page →