Skip to content
aiqbench

DeepSeek · model passport

DeepSeek V3 1 Thinking

deepseek-v3-1-thinking
  • Open weights
  • MIT licence
  • Released 2026-10-15(first seen on LMArena)
4/27axes covered
Overall · LMArena
1,416#127 of 396
Best percentile
74.0Instruction following
Cohort standing
60.9median 60.6

What the data says

Benchmarked on 4 of 27 tracked axes. Every figure below is the model's own published result on that board.

Uneven: 8.8 percentile points between its best and worst axis — 74 pct on Instruction following versus 65.2 pct on Coding.

  • Instruction following74 pct1,416Verify →
  • Hard prompts68.9 pct1,437Verify →
  • Overall68.2 pct1,416Verify →
  • Coding65.2 pct1,455Verify →
  • WebDev ArenaNot benchmarked
  • Vision ArenaNot benchmarked
  • Factuality ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • GPQA DiamondNot benchmarked
  • AIME (math)Not benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • ARC-AGI-2Not benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • LiveBenchNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

  • deepseek-v3.1-thinking stands at 60.9 (raw mean 69.1) — measured on 4 of 27 axes.

    standingtemplateVerify →

  • deepseek-v3.1-thinking stands at 60.9 versus a cohort median of 60.6 across 168 models released within 180 days — above the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →