in the loop

Qwen3-30B-A3B
#135 of 274 scored.

AlibabaApr 29, 2025Open weights

Capabilities Index

Range 129.7–138.6

136.2

Context

Takes text

41K

Input

Per 1M tokens

$0.12

Output

Per 1M tokens

$0.50

Benchmarks

How it scores.

  • GPQA Diamond48.9%

    116th of 186

  • Mock AIME62.7%

    103rd of 176

Everything else Epoch has for it

  • Lech Mazur Writing75.3%
  • Fiction.LiveBench40.6%
  • DTBench33.8%
  • WeirdML29.8%
  • LMCA18.6%
  • Chess Puzzles0.0%

On the index

Around it.

  1. #132Qwen3-30B-A3B-Instruct (Jul 2025)Alibaba137.4
  2. #133GPT-4.1OpenAI · $2.00 in136.8
  3. #134GPT-4.5OpenAI136.7
  4. #135Qwen3-30B-A3BAlibaba · $0.12 in136.2
  5. #136Qwen3-8BAlibaba136.2
  6. #137DeepSeek-V3 (Mar 2025)DeepSeek135.9
  7. #138o1-miniOpenAI135.8

Sources