Skip to content
aiqbench

Allenai · model passport

Olmo 3 32b Think

olmo-3-32b-think
  • Open weights
  • Apache 2.0 licence
  • 32.2B params
  • Released 2025-11-19(HF repo created)
8/27axes covered
Overall · LMArena
1,306#259 of 396
Best percentile
72.1LiveCodeBench (Artificial Analysis)
Cohort standing
44.1median 53.9

What the data says

Benchmarked on 8 of 27 tracked axes. Every figure below is the model's own published result on that board.

Weakest axes

  • 2.3 pct — Factuality Arena (1,319)

Uneven: 69.8 percentile points between its best and worst axis — 72.1 pct on LiveCodeBench (Artificial Analysis) versus 2.3 pct on Factuality Arena.

  • LiveCodeBench (Artificial Analysis)72.1 pct67.2%Verify →
  • IFBench (Artificial Analysis)55.6 pct49.1%Verify →
  • MMLU-Pro (Artificial Analysis)53.6 pct75.9%Verify →
  • Hard prompts39.6 pct1,326Verify →
  • Coding39.1 pct1,360Verify →
  • Instruction following37.9 pct1,298Verify →
  • Overall34.8 pct1,306Verify →
  • Factuality Arena2.3 pct1,319Verify →
  • WebDev ArenaNot benchmarked
  • Vision 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
  • 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.

How to run it

Run it yourself

Repositoryallenai/Olmo-3-32B-Think
Parameters32.2B
ArchitectureOlmo3ForCausalLM
LicenceApache 2.0

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "allenai/Olmo-3-32B-Think"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

Ollama

We don't sync Ollama tags, so we won't print a command we can't verify — a wrong one would pull the wrong model. Search the Ollama library →

Unsloth fine-tuning recipes are model-specific and we don't derive them here. Unsloth docs →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • olmo-3-32b-think stands at 44.1 (raw mean 41.9) — measured on 8 of 27 axes.

    standingtemplateVerify →

  • olmo-3-32b-think stands at 44.1 versus a cohort median of 53.9 across 128 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 →