Skip to content
aiqbench

Model passport

Openchat 3 5 0106

openchat-3-5-0106
  • Open weights
  • Apache-2.0 licence
  • Released 2024-01-06(curated)
4/27axes covered
Overall · LMArena
1,183#338 of 396
Best percentile
14.9Overall
Cohort standing
29.9median 29.6

What the data says

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

Weakest axes

  • 14.6 pct — Instruction following (1,156)
  • 14.8 pct — Coding (1,230)
  • 14.9 pct — Hard prompts (1,188)

Uneven: 0.3 percentile points between its best and worst axis — 14.9 pct on Overall versus 14.6 pct on Instruction following.

  • Overall14.9 pct1,183Verify →
  • Hard prompts14.9 pct1,188Verify →
  • Coding14.8 pct1,230Verify →
  • Instruction following14.6 pct1,156Verify →
  • 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.

  • openchat-3.5-0106 stands at 29.9 (raw mean 14.8) — measured on 4 of 27 axes.

    standingtemplateVerify →

  • openchat-3.5-0106 stands at 29.9 versus a cohort median of 29.6 across 87 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 →