Qwen2.5 VL 72B Instruct

Provider: Alibaba (Qwen) CN

Input price
$0.80
/ 1M tokens
Output price
$1.00
/ 1M tokens
Cache price
$0.40
/ 1M tokens (read)
Context
128K tokens
Max output 115K tokens

Overview

Alibaba's flagship 72B vision model handles multimodal tasks within a 128k context window.

Strengths

  • 128k context window with 115k max output
  • Recognizes common objects and analyzes layout elements
  • Supports native JSON mode output generation
  • Cached input priced at $0.4 per million

Best use cases

  • Parsing complex document charts and graphs
  • Extracting structured data from visual layouts
  • Processing image-heavy technical documentation

Pricing position

Input costs $0.8/M and output costs $1/M, making output significantly cheaper than Gemini 3.5 Flash at $4.5/M.

Caveats

No free tier is available, and the context window is smaller than newer 1M+ token siblings.

Notes drafted by an LLM from the figures on this page (2026-11-12) and rewritten when the price changes.

Key facts

Price basis
Median third-party host price (no first-party API)
Max output
115K tokens
Knowledge cutoff
2024-06-30
Listed on OpenRouter
2025-02-01
Tracked since
2026-05-08
Last updated
2026-11-10

Features

Image inputJSON output

Monthly cost estimate

Scale Input (M) Output (M) Monthly
0.1M (hobby) 0.1 0.05 $0.13
1M (small app) 1 0.5 $1.30
10M (mid scale) 10 5 $13.00
100M (large) 100 50 $130.00

※ Cache and batch discounts not factored in. Detailed calculator

Official resources

Other models from Alibaba (Qwen)

Similar-priced models

Source: OpenRouter API (qwen/qwen2.5-vl-72b-instruct)