Qwen2.5 VL 72B Instruct vs Qwen4 Max
Qwen2.5 VL 72B Instruct (Qwen) and Qwen4 Max (Qwen) are listed at different prices. Qwen2.5 VL 72B Instruct costs less: $0.80 per 1M input tokens and $1.00 per 1M output tokens, against $1.425 and $4.275 for Qwen4 Max. Qwen4 Max has the larger context window, 1,000,000 tokens against 128,000 for Qwen2.5 VL 72B Instruct. Only Qwen4 Max supports tool calling. Qwen2.5 VL 72B Instruct accepts text and images and produces text; Qwen4 Max accepts text, images and video and produces text.
Qwen: Qwen2.5 VL 72B Instruct accepts text and images, while Qwen: Qwen4 Max also accepts video; both produce text and list structured outputs and response formatting. Qwen4 Max additionally lists tool and reasoning support and has a 1M-token context, compared with 128,000 tokens for Qwen2.5 VL.
Compared from catalog data checked
Side by side
0.00 the better figure in a row| Specification | Qwen2.5 VL 72B Instruct | Qwen4 Max |
|---|---|---|
| Maker | Qwen | Qwen |
| Added to catalog | 1 Feb 2025 | 16 Nov 2026 |
| Input, per 1M tokens | $0.80 | $1.425 |
| Output, per 1M tokens | $1.00 | $4.275 |
| Cached input read, per 1M | $0.40 | $0.0713 |
| Context windowTokens | 128,000 | 1,000,000 |
| Max outputTokens | 115,200 | 131,072 |
| Accepts | image, text | image, text, video |
| Produces | text | text |
| Tool calling | No | Yes |
| Structured outputs | Yes | Yes |
| Reasoning effort control | No | Yes |
List prices through OpenRouter, in US dollars. Each model's page has its variants, price history and sample outputs.
What the same task costs
List price for one request, from the token counts shown. The same four tasks are priced on every model page.
| Task | Qwen2.5 VL 72B Instruct | Qwen4 Max |
|---|---|---|
| Answer a short question500 in / 300 out | $0.0007 | $0.0020 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.021 | $0.039 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.026 | $0.051 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.122 | $0.220 |
Which to choose
Choose Qwen2.5 VL 72B Instruct if
- Input tokens cost less: $0.80 against $1.425 per 1M
- Output tokens cost less: $1.00 against $4.275 per 1M
Choose Qwen4 Max if
- Cached input costs less: $0.0713 against $0.40 per 1M
- You need the larger context window: 1,000,000 tokens against 128,000
- You need longer answers in one response: 131,072 tokens against 115,200
- It has to read video, which the other model does not
- You need tool calling
- You want to set how much it reasons before it answers
Neither maker's quality claims are repeated here. The reliable test is your own task: ask both the same question and read the two answers.