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

Qwen3 Next 80B A3B Instruct vs Qwen3 VL 235B A22B Instruct

Alibaba · released

Wins 2 of 6 areas

Reasoning · Long documents

Alibaba · released

Wins 2 of 6 areas

Facts · Following instructions

The two are evenly matched.Qwen3 Next 80B A3B Instruct is better at reasoning and long documents; Qwen3 VL 235B A22B Instruct at facts and following instructions.

Scores updated · 27 tests both models report · How we compare

Where each one wins

Tests won in each of the six 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 Next 80B A3B Instruct$0.15 to read · $1.20 to write$1.35Price from Alibaba's own price page
Qwen3 VL 235B A22B Instruct$0.40 to read · $1.60 to write$2.00Price from Alibaba's own price page

Qwen3 Next 80B A3B Instruct costs 33% 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 Next 80B A3B Instruct pulls ahead

  • Reasons across sets of long documentsAA-LCR+20points aheadQwen3 Next 80B A3B Instruct52.7Qwen3 VL 235B A22B Instruct32.7
  • Graduate-level biology, physics and chemistry questionsGPQA Diamond+2.6points aheadQwen3 Next 80B A3B Instruct73.8Qwen3 VL 235B A22B Instruct71.2
  • Recent programming contest problemsLiveCodeBench v6+2.3points aheadQwen3 Next 80B A3B Instruct56.6Qwen3 VL 235B A22B Instruct54.3

Where Qwen3 VL 235B A22B Instruct pulls ahead

  • Code for real scientific research problemsSciCode+5.2points aheadQwen3 VL 235B A22B Instruct35.9Qwen3 Next 80B A3B Instruct30.7
  • Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+3points aheadQwen3 VL 235B A22B Instruct20.3Qwen3 Next 80B A3B Instruct17.3
  • Follows unfamiliar, precisely checkable instructionsIFBench+3points aheadQwen3 VL 235B A22B Instruct42.7Qwen3 Next 80B A3B Instruct39.7

Every test, side by side

All 27 tests both models report. The winning score is in its model's colour; marks a score checked independently.

CodingEven

Full coding ranking

AgentsEven

Full agents ranking

ReasoningQwen3 Next 80B A3B Instruct

Full reasoning ranking

FactsQwen3 VL 235B A22B Instruct

Full facts ranking

Long documentsQwen3 Next 80B A3B Instruct
  • AA-LCRQwen3 Next 80B A3B Instruct by 2052.732.7+20

Full long documents ranking

Following instructionsQwen3 VL 235B A22B Instruct
  • IFBenchQwen3 VL 235B A22B Instruct by 339.742.7+3

Full following instructions ranking

Other results16 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 Next 80B A3B Instruct or Qwen3 VL 235B A22B Instruct?

Qwen3 Next 80B A3B Instruct and Qwen3 VL 235B A22B Instruct each win two of the six areas where both have results. Qwen3 Next 80B A3B Instruct wins reasoning and long documents; Qwen3 VL 235B A22B Instruct wins facts and following instructions. Qwen3 Next 80B A3B Instruct costs 33% less. They are level on coding and agents.

Which is better for coding?

Neither. They win 1 coding test each of the 3 both models report, and 1 is a tie.

Which is cheaper?

Qwen3 Next 80B A3B Instruct costs $0.15 per million input tokens and $1.20 per million output tokens; Qwen3 VL 235B A22B Instruct costs $0.40 and $1.60. That makes Qwen3 Next 80B A3B Instruct about 33% cheaper for the same work.

How do you compare the two?

We use the 27 benchmark tests both models have published scores on. The verdict counts the 11 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 16 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.