Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d271Policy30d66, 14 pendingSources1,015

Qwen3 VL 235B A22B Reasoning vs Qwen3 VL 8B Instruct

Alibaba · released

Wins 7 of 7 areas

Coding · Agents · Reasoning · Facts · Images and charts · Long documents · Following instructions

Alibaba · released

Wins 0 of 7 areas

—

Qwen3 VL 235B A22B Reasoning is the stronger all-rounder.Qwen3 VL 8B Instruct is cheaper.

Scores updated · 47 tests both models report · How we compare

Where each one wins

Tests won in each of the seven 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, long documents and following instructions rest on a single test each.

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.

Qwen3 VL 235B A22B Reasoning$0.40 to read · $4.00 to write$4.40Price from Alibaba's own price page
Qwen3 VL 8B Instruct$0.18 to read · $0.70 to write$0.88Price from Alibaba's own price page

Qwen3 VL 8B Instruct costs 80% less for the same work.

The biggest differences

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

Where Qwen3 VL 235B A22B Reasoning pulls ahead

  • Reasons across sets of long documentsAA-LCR+47points aheadQwen3 VL 235B A22B Reasoning63.7Qwen3 VL 8B Instruct16.7
  • Graduate-level biology, physics and chemistry questionsGPQA Diamond+34.5points aheadQwen3 VL 235B A22B Reasoning77.2Qwen3 VL 8B Instruct42.7
  • Recent programming contest problemsLiveCodeBench v6+30.8points aheadQwen3 VL 235B A22B Reasoning70.1Qwen3 VL 8B Instruct39.3

Where Qwen3 VL 8B Instruct pulls ahead

  • Reads text in imagesOCRBench+2.1points aheadQwen3 VL 8B Instruct89.6Qwen3 VL 235B A22B Reasoning87.5
  • Visual perception tasks people solve at a glanceBLINK+2points aheadQwen3 VL 8B Instruct69.1Qwen3 VL 235B A22B Reasoning67.1

Every test, side by side

All 47 tests both models report. The winning score is in its model's colour; marks a score checked independently.

CodingQwen3 VL 235B A22B Reasoning

Full coding ranking

AgentsQwen3 VL 235B A22B Reasoning

Full agents ranking

ReasoningQwen3 VL 235B A22B Reasoning

Full reasoning ranking

FactsQwen3 VL 235B A22B Reasoning

Full facts ranking

Images and chartsQwen3 VL 235B A22B Reasoning
  • MMMU-ProQwen3 VL 235B A22B Reasoning by 21.468.747.3+21.4
  • CharXiv (RQ)Qwen3 VL 235B A22B Reasoning by 19.766.146.4+19.7
  • MathVistaQwen3 VL 235B A22B Reasoning by 8.685.877.2+8.6
  • MMStarQwen3 VL 235B A22B Reasoning by 7.878.770.9+7.8
  • Video-MMEQwen3 VL 235B A22B Reasoning by 7.67971.4+7.6
  • OCRBenchQwen3 VL 8B Instruct by 2.187.589.6+2.1
  • BLINKQwen3 VL 8B Instruct by 267.169.1+2

Full images and charts ranking

Long documentsQwen3 VL 235B A22B Reasoning
  • AA-LCRQwen3 VL 235B A22B Reasoning by 4763.716.7+47

Full long documents ranking

Following instructionsQwen3 VL 235B A22B Reasoning
  • IFBenchQwen3 VL 235B A22B Reasoning by 24.256.532.3+24.2

Full following instructions ranking

Other results29 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, Qwen3 VL 235B A22B Reasoning or Qwen3 VL 8B Instruct?

Qwen3 VL 235B A22B Reasoning wins all seven areas where both have results: coding, agents, reasoning, facts, images and charts, long documents and following instructions. Qwen3 VL 8B Instruct wins none, but costs 80% less.

Which is better for coding?

Qwen3 VL 235B A22B Reasoning. It wins 3 of the 3 coding tests both models report; Qwen3 VL 8B Instruct wins none.

Which is cheaper?

Qwen3 VL 235B A22B Reasoning costs $0.40 per million input tokens and $4.00 per million output tokens; Qwen3 VL 8B Instruct costs $0.18 and $0.70. That makes Qwen3 VL 8B Instruct about 80% cheaper for the same work.

How do you compare the two?

We use the 47 benchmark tests both models have published scores on. The verdict counts the 18 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 29 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.