Qwen3 VL Thinking (8B) is the stronger all-rounder.
Scores updated · 15 tests both models report · How we compare
Where each one wins
Tests won in each of the one area 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.
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.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Qwen3 VL Thinking (8B) pulls ahead
Where InternVL-3.5 (8B) pulls ahead
- Reads text in imagesOCRBench+1.8points ahead
Every test, side by side
All 15 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Images and chartsQwen3 VL Thinking (8B)
Other results10 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.
- ScreenSpot-Pro (No tools)Qwen3 VL Thinking (8B) by 31.215.446.6+31.2
- OSWorld-GQwen3 VL Thinking (8B) by 24.831.956.7+24.8
- MATH-VisionQwen3 VL Thinking (8B) by 10.65262.7+10.6
- ScreenSpot-V2Qwen3 VL Thinking (8B) by 9.68493.6+9.6
- PhyXQwen3 VL Thinking (8B) by 7.250.557.7+7.2
- ReMIQwen3 VL Thinking (8B) by 4.552.657.2+4.5
- AI2DQwen3 VL Thinking (8B) by 2.682.384.9+2.6
- HumanEval-VQwen3 VL Thinking (8B) by 2.624.326.9+2.6
- MMBench (EN)Qwen3 VL Thinking (8B) by 2.488.290.5+2.4
- All-Angles-Benchtie45.345.9tie
Questions people ask
Which is better, InternVL-3.5 (8B) or Qwen3 VL Thinking (8B)?
Qwen3 VL Thinking (8B) wins the one area where both have results: images and charts. InternVL-3.5 (8B) wins none.
How do you compare the two?
We use the 15 benchmark tests both models have published scores on. The verdict counts the 5 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 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.