Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d270Policy30d65, 13 pendingSources1,015

Qwen2.5 VL 72B vs Qwen3 VL 30B A3B Instruct

Alibaba · released

Wins 0 of 2 areas

—

Alibaba · released

Wins 2 of 2 areas

Agents · Images and charts

Qwen3 VL 30B A3B Instruct is the stronger all-rounder.

Scores updated · 17 tests both models report · How we compare

Where each one wins

Tests won in each of the two areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.

Agents rests on a single test.

What it costs

Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.

Qwen2.5 VL 72BNo current price is tracked.
Qwen3 VL 30B A3B Instruct$0.20 to read · $0.80 to write$1.00Price from Alibaba's own price page

The biggest differences

The tests each model wins by the widest margin, up to three each. Scores are out of 100.

Where Qwen3 VL 30B A3B Instruct pulls ahead

  • Completes tasks by operating a computer desktopOSWorld-Verified+21.5points aheadQwen3 VL 30B A3B Instruct30.3Qwen2.5 VL 72B8.8
  • Harder college exam questions with imagesMMMU-Pro+11points aheadQwen3 VL 30B A3B Instruct62.1Qwen2.5 VL 72B51.1
  • Math problems shown in pictures and chartsMathVista+5.3points aheadQwen3 VL 30B A3B Instruct80.1Qwen2.5 VL 72B74.8

Where Qwen2.5 VL 72B pulls ahead

No clear win on a test scored out of 100.

Every test, side by side

All 17 tests both models report. The winning score is in its model's colour; marks a score checked independently.

AgentsQwen3 VL 30B A3B Instruct

Full agents ranking

Images and chartsQwen3 VL 30B A3B Instruct
  • MMMU-ProQwen3 VL 30B A3B Instruct by 1151.162.1+11
  • MathVistaQwen3 VL 30B A3B Instruct by 5.374.880.1+5.3
  • BLINKQwen3 VL 30B A3B Instruct by 4.76367.7+4.7
  • OCRBenchQwen3 VL 30B A3B Instruct by 1.888.590.3+1.8
  • MMStarQwen3 VL 30B A3B Instruct by 1.370.872.1+1.3

Full images and charts ranking

Other results11 tests, not counted

Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.

Questions people ask

Which is better, Qwen2.5 VL 72B or Qwen3 VL 30B A3B Instruct?

Qwen3 VL 30B A3B Instruct wins all two areas where both have results: agents and images and charts. Qwen2.5 VL 72B wins none.

How do you compare the two?

We use the 17 benchmark tests both models have published scores on. The verdict counts the 6 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 11 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.