modelbenchmark.io

Qwen2.5 72B

Alibaba · qwen-2-5-72b-instruct

Compare

Qwen instruction model for multilingual chat, reasoning, and tool use

Specification

most-agreed values

Context
33K
Max output
16K
Released
2025-07-03
Knowledge cutoff
2024-04
Retires
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.36
Output
$0.40
Cache read
$0.1785

Quality

3 benchmarks · 1 source

Composite
33rd
percentile of 334 scored models
Rank
225 / 334
±0.06 sd
Evidence
3 × 1
one source, or fewer than 3 benchmarks
Effort range
one configuration only
reasoning32nd1/4 bench
math36th2/7 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
MATH level 563.2 ±1.1defaultEpoch AI2025-01-27
GPQA diamond49.1 ±2.7defaultEpoch AI2025-01-27
OTIS Mock AIME 2024-20258.1 ±2.6defaultEpoch AI2025-02-25

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 5 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
nano-gpt$0.357$0.408$0.1785131K8K
kilo$0.36$0.4033K16K
openrouter$0.36$0.4033K16K
novita$0.38$0.4032K8K
novita-ai$0.38$0.4032K8K

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02openrouterInput$0.36
2026-09-06openrouterInput$0.36
2026-09-02openrouterOutput$0.40
2026-09-06openrouterOutput$0.40

Other listings of this model

same model, different key

More from Alibaba

most-hosted first