Qwen3 30B A3B 2507 (Reasoning)
Jul 2025
Qwen3 30B A3B 2507 (Reasoning)Compare inon
Benchmarks of Qwen3 30B A3B 2507 (Reasoning)30
BenchmarkByScoreRank
Overall1
Knowledge4
Science3
Coding3
Agents4
Health & law1
Documents2
Legacy · saturated or no longer reported12
No benchmark matches
$0.066per task?
price not verifiedprice
—output speed?
262Kcontext
Rankingin thinking mode?
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGPT-5 Minihigh+19%-19%$0.053BetterGPT OSS 20BhighOpen+10%-84%$0.011Context: 262K → 131KSOpenApriel v1.6 15B Thinker+14%——Context: 262K → 128K
CheaperNemotron 3 Nano 30B A3BthinkingOpen+5%-75%$0.017Context: 262K → 131K+1BetterQwen3 235B A22B Thinking 2507thinkingOpen+12%+21%$0.08Memory (4-bit): 18 GB → 141 GB+1ModelScoreCostPer taskWhat changes
GPT-5 Minihigh+19%-19%$0.053GPT OSS 20BhighOpen+10%-84%$0.011Context: 262K → 131KNemotron 3 Nano 30B A3BthinkingOpen+5%-75%$0.017Context: 262K → 131K+1ModelScoreCostPer taskWhat changes
GPT OSS 20BhighOpen+10%-84%$0.011Context: 262K → 131KNemotron 3 Nano 30B A3BthinkingOpen+5%-75%$0.017Context: 262K → 131K+1GPT-5 Minihigh+19%-19%$0.053Qwen3 235B A22B Thinking 2507thinkingOpen+12%+21%$0.08Memory (4-bit): 18 GB → 141 GB+1GPT OSS 120BhighOpen+24%+63%$0.11Context: 262K → 131K+1ModelScoreCostPer taskWhat changes
SApriel v1.6 15B ThinkerOpen+14%——Context: 262K → 128KGPT OSS 20BhighOpen+10%-84%$0.011Context: 262K → 131K
Seed OSS 36B InstructthinkingOpen+8%—$0.21 / $0.57Memory (4-bit): 18 GB → 22 GB
EXAONE 4.0 32BthinkingOpen+6%——Context: 262K → 131K+1Nemotron 3 Nano 30B A3BthinkingOpen+5%-75%$0.017Context: 262K → 131K+1ModelScoreCostPer taskWhat changes
Qwen3 235B A22B Thinking 2507thinkingOpen+12%+21%$0.08Memory (4-bit): 18 GB → 141 GB+1Qwen3 Next 80B A3B ThinkingthinkingOpen+11%—$0.50 / $6Memory (4-bit): 18 GB → 49 GBQwen3 Maxnon-thinking+9%—$1.20 / $6Qwen3 VL 32B (Reasoning)Open+4%—$0.20 / $1.50Context: 262K → 256K+1Qwen3 VL 30B A3B ThinkingthinkingOpensame—$0.29 / $0.99Context: 262K → 256K+1Qwen3 Next 80B A3B Instructnon-thinkingOpen-3%—$0.50 / $2Memory (4-bit): 18 GB → 49 GBQwen3 Omni 30B A3B (Reasoning)Open-4%—$0.25 / $0.97Context: 262K → 66K+1Qwen3 Max (Preview)non-thinking-8%—$0.87 / $3.48All Alibaba (Qwen) models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✓ 24 GB GPU · compressed✓ 48 GB GPU · 8-bit✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
History
2 versionsBenchmarks
all categoriesChat & Knowledge
Humanity's Last Exam10.3#189
AA Omniscience Index-56.1#244
CritPt (physics)0.3#151
AA Intelligence Index 4.3.29.8#213
AA Omniscience · Accuracy16.3#217
AA-LCR (long context) 1.161.3#155
GDP.pdf · AA all-pass0.4#130
MLCR-AA2.2#90
MMLU-Pro · official boardlab80.9
GPQA Diamond70.7#182
AIME 2025 · AA archived56.3#84
Coding
SciCode33#120
Terminal-Bench 2.11.5#148
LiveCodeBench · AA archived70.7#44
Agents & Tool Use
AutomationBench-AA0.8#116
GDPval-AA (work tasks) 2.167.1#122
τ³-Banking · AA5.4#116
AA-Briefcase 1.1 · Elo5.1#130
τ²-bench · AA28.1#154
Writing & Language
IFBench (instructions)50.7#114
FAQ
How much does Qwen3 30B A3B 2507 (Reasoning) cost?›
There is no per-unit API price: price not verified.
How good is Qwen3 30B A3B 2507 (Reasoning), and what is it best at?›
#188 of 336 in Chat & Knowledge, #77 of 134 in Coding, #116 of 135 in Agents & Tool Use, #113 of 245 in Writing & Language, #130 of 135 in Vision & Documents.
Is Qwen3 30B A3B 2507 (Reasoning) better than Qwen3 14B?›
On LiveCodeBench · AA archived, Qwen3 30B A3B 2507 (Reasoning) scores 70.7 vs 52.3 — higher.
What is the context window of Qwen3 30B A3B 2507 (Reasoning)?›
262K tokens.
When was Qwen3 30B A3B 2507 (Reasoning) released? Is there a newer version?›
Released 2025-07-30. It has been superseded by Qwen 3.5 35B A3B.
Is Qwen3 30B A3B 2507 (Reasoning) open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Qwen3 30B A3B 2507 (Reasoning) locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Qwen3 30B A3B 2507 (Reasoning) read images?›
No, it accepts text only.
What is Qwen3 30B A3B 2507 (Reasoning)?›
Qwen3 30B A3B 2507 (Reasoning) is an open-weights language model from Alibaba (Qwen), released Jul 2025, the successor to Qwen3 14B.
What are the best alternatives to Qwen3 30B A3B 2507 (Reasoning)?›
GPT-5 Mini (cheaper), GPT OSS 20B (better), Apriel v1.6 15B Thinker (open weights).