Qwen2.5 72B Instruct

Alibaba llm

Alibaba's flagship dense instruction-tuned model in the Qwen2.5 series, which spans 0.5B to 72B parameters.

overall ⌄

32.1

85 of 93 ranked

price · $/M tokens

self-hosted

open weights, no first-party API

to run it yourself ⌄

43 GB

minimum · 173 GB recommended

Benchmarks

What its maker published, and what anyone else measured. Every figure links the document it came from.

benchmark	published	measured
MMLU-Pro	71.1	—
GPQA	49	—

No recorded lineage — root or standalone model.