Skip to content
aiqbench

Ai21-Labs · model passport

Jamba 1 5 Mini

jamba-1-5-mini
  • Open weights
  • Jamba Open licence
  • Released 2024-08-22(curated)
6/27axes covered
Overall · LMArena
1,240#305 of 396
Best percentile
23.2Overall
Cohort standing
27.6median 37.6

What the data says

Benchmarked on 6 of 27 tracked axes. Every figure below is the model's own published result on that board.

Weakest axes

  • 5 pct — LiveCodeBench (Artificial Analysis) (6.2%)
  • 7.5 pct — MMLU-Pro (Artificial Analysis) (37.1%)

Uneven: 18.2 percentile points between its best and worst axis — 23.2 pct on Overall versus 5 pct on LiveCodeBench (Artificial Analysis).

  • Overall23.2 pct1,240Verify →
  • Hard prompts22 pct1,237Verify →
  • Instruction following20.7 pct1,205Verify →
  • Coding20.2 pct1,264Verify →
  • MMLU-Pro (Artificial Analysis)7.5 pct37.1%Verify →
  • LiveCodeBench (Artificial Analysis)5 pct6.2%Verify →
  • 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
  • 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.

  • jamba-1.5-mini stands at 27.6 (raw mean 16.4) — measured on 6 of 27 axes.

    standingtemplateVerify →

  • jamba-1.5-mini stands at 27.6 versus a cohort median of 37.6 across 126 models released within 180 days — below 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 →