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

Qwen3 VL 235B A22B Instruct vs Qwen3 VL 30B A3B Reasoning

Alibaba · released

Wins 2 of 7 areas

Coding · Images and charts

Alibaba · released

Wins 3 of 7 areas

Reasoning · Long documents · Following instructions

Qwen3 VL 30B A3B Reasoning wins more areas, narrowly.Qwen3 VL 235B A22B Instruct is cheaper and better at images and charts.

Scores updated · 51 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.

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 Instruct$0.40 to read · $1.60 to write$2.00Price from Artificial Analysis
Qwen3 VL 30B A3B Reasoning$0.20 to read · $2.40 to write$2.60Price from Artificial Analysis

Qwen3 VL 235B A22B Instruct costs 23% less for the same work.

The biggest differences

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

Where Qwen3 VL 30B A3B Reasoning pulls ahead

  • Reasons across sets of long documentsAA-LCR+10points aheadQwen3 VL 30B A3B Reasoning42.7Qwen3 VL 235B A22B Instruct32.7
  • Recent programming contest problemsLiveCodeBench v6+9.9points aheadQwen3 VL 30B A3B Reasoning64.2Qwen3 VL 235B A22B Instruct54.3
  • Follows unfamiliar, precisely checkable instructionsIFBench+2.4points aheadQwen3 VL 30B A3B Reasoning45.1Qwen3 VL 235B A22B Instruct42.7

Where Qwen3 VL 235B A22B Instruct pulls ahead

  • Completes tasks by operating a computer desktopOSWorld-Verified+36.1points aheadQwen3 VL 235B A22B Instruct66.7Qwen3 VL 30B A3B Reasoning30.6
  • Reads text in imagesOCRBench+8.1points aheadQwen3 VL 235B A22B Instruct92Qwen3 VL 30B A3B Reasoning83.9
  • Code for real scientific research problemsSciCode+7.1points aheadQwen3 VL 235B A22B Instruct35.9Qwen3 VL 30B A3B Reasoning28.8

Every test, side by side

All 51 tests both models report. The winning score is in its model's colour; marks a score checked independently.

CodingQwen3 VL 235B A22B Instruct

Full coding ranking

AgentsEven
  • OSWorld-VerifiedQwen3 VL 235B A22B Instruct by 36.166.730.6+36.1
  • GDPValQwen3 VL 30B A3B Reasoning by 1.66.78.3+1.6

Full agents ranking

ReasoningQwen3 VL 30B A3B Reasoning

Full reasoning ranking

FactsEven

Full facts ranking

Images and chartsQwen3 VL 235B A22B Instruct
  • OCRBenchQwen3 VL 235B A22B Instruct by 8.19283.9+8.1
  • MMMU-ProQwen3 VL 235B A22B Instruct by 5.867.661.8+5.8
  • CharXiv (RQ)Qwen3 VL 235B A22B Instruct by 5.562.156.6+5.5
  • BLINKQwen3 VL 235B A22B Instruct by 5.370.765.4+5.3
  • MathVistaQwen3 VL 235B A22B Instruct by 384.981.9+3
  • MMStarQwen3 VL 235B A22B Instruct by 2.978.475.5+2.9

Full images and charts ranking

Long documentsQwen3 VL 30B A3B Reasoning
  • AA-LCRQwen3 VL 30B A3B Reasoning by 1032.742.7+10

Full long documents ranking

Following instructionsQwen3 VL 30B A3B Reasoning
  • IFBenchQwen3 VL 30B A3B Reasoning by 2.442.745.1+2.4

Full following instructions ranking

Other results33 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 Instruct or Qwen3 VL 30B A3B Reasoning?

Qwen3 VL 30B A3B Reasoning wins three of the seven areas where both have results: reasoning, long documents and following instructions. Qwen3 VL 235B A22B Instruct wins coding and images and charts, and costs 23% less. They are level on agents and facts.

Which is better for coding?

Qwen3 VL 235B A22B Instruct. It wins 2 of the 3 coding tests both models report; Qwen3 VL 30B A3B Reasoning wins 1.

Which is cheaper?

Qwen3 VL 235B A22B Instruct costs $0.40 per million input tokens and $1.60 per million output tokens; Qwen3 VL 30B A3B Reasoning costs $0.20 and $2.40. That makes Qwen3 VL 235B A22B Instruct about 23% cheaper for the same work.

How do you compare the two?

We use the 51 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 33 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.