Skip to content
Model Pulse

Model

Qwen2.5-VL 72B Instruct

Qwen vision-language model for visual reasoning, documents, and agent tasks

9 providers · 10 listingsreleased knowledge cutoff 2024-04open weights
$0.7
lowest input /M · NanoGPT
$0.7
lowest output /M · NanoGPT
131K
context window
115K
max output

Compare providers (sorted by listed input price)

8 of 9 providers publish a price for Qwen2.5-VL 72B Instruct. The cheapest full-context option is Kilo Gateway at $0.8/$1 per M — 81% below Alibaba's own price. NanoGPT lists lower but caps context at 32K.

10 listings
Cache read /MCache write /MCapabilitiesStatus
NanoGPTvariant
qwen25-vl-72b-instruct
$0.7$0.7$0.35—32Kno reasoningno toolsno structuredvisionMay 10, 2025
NanoGPTvariant
TEE/qwen2.5-vl-72b-instruct
$0.7$0.7$0.35—66Kno reasoningno toolsno structuredvisionFeb 1, 2025
NovitaAI
qwen/qwen2.5-vl-72b-instruct
$0.8$0.8——33Kno reasoningno toolsno structuredvisionMar 25, 2025
Kilo Gateway
qwen/qwen2.5-vl-72b-instruct
$0.8$1$0.4—128Kno reasoningno toolsstructuredvisionFeb 1, 2025
OpenRouter
qwen/qwen2.5-vl-72b-instruct
$0.8$1$0.4—128Kno reasoningno toolsstructuredvisionFeb 1, 2025
OVHcloud AI Endpoints
qwen2.5-vl-72b-instruct
$1.01$1.01——33Kno reasoningno toolsstructuredvisionMar 31, 2025
Cortecs
qwen2.5-vl-72b-instruct
$1.01$1.01——32Kreasoningno toolsstructuredvisionJan 27, 2025
Alibaba (China)
qwen2-5-vl-72b-instruct
$2.29$6.88——131Kno reasoningtoolsno structuredno visionSep 1, 2024
Alibaba
qwen2-5-vl-72b-instruct
$2.80$8.40——131Kno reasoningtoolsno structuredno visionSep 1, 2024
Qiniu
qwen2.5-vl-72b-instruct
————128Kno reasoningtoolsno structuredvisionAug 5, 2025
1–10 of 10 listings

Best price history

Daily lowest published price across live providers. A lower line is better for the buyer.

Best input /M
$0.7▲ 180.0%
Best output /M
$0.7▼ 6.3%
Provider coverage
10
13 at the start · 24 snapshots
Best published input and output price historyDaily lowest published price per one million tokens. Accent is input price and violet is output price. The vertical scale starts at zero.$0$0.405$0.81Aug 21, 2026Sep 1, 2026Sep 14, 2026
Best input priceBest output priceAug 21, 2026 → Sep 14, 2026

Price badge

Current price badge for Qwen2.5-VL 72B Instruct
![Qwen2.5-VL 72B Instruct price](https://www.vaanalytics.in/badge/alibaba/qwen2-5-vl-72b-instruct.svg)
https://www.vaanalytics.in/badge/alibaba/qwen2-5-vl-72b-instruct.svg

Live SVG, regenerated on every hourly sync — paste it into a README to always show the current cheapest listed price. Updates when prices change; no build step needed on your side.

Recent changes

Changelog →
  • contextvia nano-gpt · Sep 5, 2026
    • context: 128K → 32K tokens
    • input: 128K → 32K tokens
  • repricedvia cortecs · Sep 5, 2026
    • input: $0.25 → $1.01
    • output: $0.747 → $1.01
  • contextvia nano-gpt · Sep 3, 2026
    • context: 32K → 128K tokens
    • output: 33K → 115K tokens
    • input: 32K → 128K tokens
  • repricedvia openrouter · Sep 3, 2026
    • output: 29K → 115K tokens
    • input: $0.25 → $0.8
    • output: $0.75 → $1
    • cache_read: — → $0.4
  • repricedvia kilo · Sep 3, 2026
    • context: 32K → 128K tokens
    • output: 29K → 115K tokens
    • input: $0.25 → $0.8
    • output: $0.75 → $1
    • +1 more
  • repricedvia kilo · Aug 26, 2026
    • context: 128K → 32K tokens
    • output: 115K → 29K tokens
    • input: $0.8 → $0.25
    • output: $1 → $0.75
    • +1 more

Weights

Licence unclearother73B params

No machine-readable licence on the model card. Check it before you rely on this. Weights download without an access request.

licence and gating from the Hugging Face model card ↗ — the card is authoritative, this is a summary

Prices are per million tokens (USD). “—” means the provider does not publicly list a price for this model. Data from models.dev; verify with the provider before purchasing.