modelbenchmark.io

qwen2.5-vl-72b-instruct

Alibaba · qwen2-5-vl-72b-instruct

Compare

Qwen2.5-VL is a powerful vision-language model with advanced capabilities in visual understanding, long video reasoning, and structured output generation.

Specification

most-agreed values

Context
32K
Max output
32K
Released
2025-01-26
Knowledge cutoff
2024-04
Retires
2026-01-05
Open weights
yes
Input
text, image
Output
text

Price

US dollars per million tokens · most-agreed

Input
$1.014
Output
$1.014
Cache read
$0.35

Available from 12 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
nebius$0.25$0.75——32K32K—
nano-gpt$0.6999$0.6999——32K32K—
nano-gpt · tee$0.70$0.70$0.35—66K8K—
kilo$0.80$1.00$0.40—128K115K—
novita$0.80$0.80——33K33K—
openrouter$0.80$1.00$0.40—128K115K—
fireworks_ai$0.90$0.90——128K128K—
ovhcloud · ovhcloud$0.91$0.91——32K32K—
ovhcloud$1.01$1.01——33K33K—
cortecs$1.014$1.014——32K32K—
together_ai$1.95$8.00——33K33K2026-01-05
qiniu-ai$2.3188$6.9565——128K8K—

Price history

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

DateHostField$/M
2026-09-02cortecsInput$0.25
2026-09-05cortecsInput$1.014
2026-09-02kiloInput$0.25
2026-09-04kiloInput$0.80
2026-09-02nebiusInput$0.25
2026-09-03nebiusInput$0.13
2026-09-06nebiusInput$0.25
2026-09-02openrouterInput$0.25
2026-09-04openrouterInput$0.80
2026-09-06openrouterInput$0.80
2026-10-10openrouterInput$0.80
2026-09-02cortecsOutput$0.747
2026-09-05cortecsOutput$1.014
2026-09-02kiloOutput$0.75
2026-09-04kiloOutput$1.00
2026-09-02nebiusOutput$0.75
2026-09-03nebiusOutput$0.40
2026-09-06nebiusOutput$0.75
2026-09-02openrouterOutput$0.75
2026-09-04openrouterOutput$1.00
2026-09-06openrouterOutput$1.00
2026-10-10openrouterOutput$1.00
2026-09-04openrouterCache read$0.40
2026-09-06openrouterCache read$0.40
2026-10-10openrouterCache read$0.40
2026-09-15together_aiInput$1.95
2026-09-24together_aiInput$1.95
2026-09-15together_aiOutput$8.00
2026-09-24together_aiOutput$8.00

More from Alibaba

most-hosted first