Skip to content
aiqbench

Alibaba · model passport

Qwen2 5 72b Instruct

qwen2-5-72b-instruct
  • Open weights
  • Qwen licence
  • 72.7B params
  • Released 2024-09-19(curated)
7/27axes covered
Overall · LMArena
1,303#265 of 396
Best percentile
48.8LiveBench
Cohort standing
39.9median 38.1

What the data says

Benchmarked on 7 of 27 tracked axes. Every figure below is the model's own published result on that board.

Uneven: 25.3 percentile points between its best and worst axis — 48.8 pct on LiveBench versus 23.5 pct on AIME (math).

  • LiveBench48.8 pct52.9%Verify →
  • Coding37.3 pct1,355Verify →
  • Instruction following36.1 pct1,292Verify →
  • Hard prompts35.9 pct1,317Verify →
  • GPQA Diamond33.9 pct49.1%Verify →
  • Overall33.3 pct1,303Verify →
  • AIME (math)23.5 pct8.1%Verify →
  • WebDev ArenaNot benchmarked
  • Vision ArenaNot benchmarked
  • Factuality ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot 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.

How to run it

Run it yourself

RepositoryQwen/Qwen2.5-72B-Instruct
Parameters72.7B
ArchitectureQwen2ForCausalLM
LicenceQwen

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "Qwen/Qwen2.5-72B-Instruct"
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.

  • qwen2.5-72b-instruct stands at 39.9 (raw mean 35.6) — measured on 7 of 27 axes.

    standingtemplateVerify →

  • qwen2.5-72b-instruct stands at 39.9 versus a cohort median of 38.1 across 133 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 →