Sign in

Is DeepSeek V4 Pro Preview SOTA in Long context?#66 in Long contextRanked #66 in Long context.

How it scores here

DeepSeek V4 Pro Preview in Long context

Primary benchmarks

  • LMArena Text (style-controlled, longer query): 1464 (CI 1459–1470) on Nov 19, 2026 (#82 of 248).

On the category's weighted score it is 39.4 points behind #1 Gemini 4 Argon and 0.6 behind the next model up, Ernie 5.1.

Nearby on the board

#ModelCategory score
63Grok 4.20 Beta62.4
64Grok 4.20 Beta 161.8
65Ernie 5.161.2
66DeepSeek V4 Pro Preview60.6
67Qwen3 7 Plus60.0
68Gemini 3.5 Flash Lite59.4
69Step 5 Preview58.8
Standings

Long context

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

Category page →