Skip to content
aiqbench

DeepSeek · model passport

DeepSeek Coder V2

deepseek-coder-v2
  • Open weights
  • DeepSeek License licence
  • Released 2024-06-17(curated)
5/27axes covered
Overall · LMArena
1,265#294 of 396
Best percentile
35.7LiveBench
Cohort standing
37.9median 36.2

What the data says

Benchmarked on 5 of 27 tracked axes. Every figure below is the model's own published result on that board.

Uneven: 9.7 percentile points between its best and worst axis — 35.7 pct on LiveBench versus 26 pct on Overall.

  • LiveBench35.7 pct46.3%Verify →
  • Coding34 pct1,342Verify →
  • Hard prompts30.3 pct1,288Verify →
  • Instruction following27 pct1,253Verify →
  • Overall26 pct1,265Verify →
  • 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
  • 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-coder-v2 stands at 37.9 (raw mean 30.6) — measured on 5 of 27 axes.

    standingtemplateVerify →

  • deepseek-coder-v2 stands at 37.9 versus a cohort median of 36.2 across 125 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 →