Benchmark Almanac

Qwen3-30B-A3B-Thinking Jul 2025

Qwen3-30B-A3B-Thinking.

Qwen Released 30 Jul 2025
17
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Qwen Jul 2025

Qwen3-30B-A3B-Thinking

Jul 2025 snapshot Open weights

139.6 #127 of 271
Epoch Capabilities Index
  • CritPt 0.29% #109 /160
  • MathArena (Apex, IMO, USAMO) 0.52% #42 /47
  • GPQA Diamond 70.1% #109 /219

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

BenchmarkScoreRankSource
MMLU-Pro 80.9% #53 of 212 Official leaderboard
Vectara Hallucination Leaderboard (HHEM) 4.80% #61 of 148 Official leaderboard
LongBench v2 50.1% #15 of 33 Official leaderboard

Math and science

BenchmarkScoreRankSource
GPQA Diamond 70.1% #109 of 219 Epoch AI
OTIS Mock AIME 2024-2025 70.3% #101 of 202 Epoch AI
HiPhO 54.6% #18 of 35 Official leaderboard
DTBench 48.9% #112 of 178 Official leaderboard
CritPt 0.29% #109 of 160 Epoch AI
MathArena (Apex, IMO, USAMO) 0.52% #42 of 47 Official leaderboard
HMMT 2026 78.8% #28 of 31 Official leaderboard
AIME 2026 88.3% #29 of 31 Official leaderboard

Agents and tools

BenchmarkScoreRankSource
LLM Chess -37 #66 of 124 Official leaderboard

Professional work

BenchmarkScoreRankSource
StockBench 2.10% #5 of 13 Official leaderboard
ProfBench 37.6% #67 of 72 Official leaderboard

Other

BenchmarkScoreRankSource
ACCoRD 57.0 #97 of 150 Official leaderboard
Conceptual Reasoning Index 32.9 #109 of 149 Official leaderboard
LMCA 19.5% #105 of 139 Epoch AI

Qwen's models over time

Qwen's models in color, against every other model on the Epoch Capabilities Index, by release date.

50 75 100 125 150 175 2023 2024 2025 2026 2027 Qwen3.8 Max (0902) Qwen3.7 Flash Qwen3-Max Qwen3-30B-A3B Qwen2.5-72B Qwen2.5-7B Qwen2.5-Coder (1.5B) 50 75 100 125 150 175 2023 2024 2025 2026 2027 Qwen3.8 Max (0902) Qwen3.7 Flash