Qwen3 32B
Qwen3 32B is behind the leaders in factuality. Too few results yet to rate reasoning, coding, agentic tasks, safety, long context, math, multimodal tasks, multilingual tasks, or instruction following.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Reasoning, Coding, Agentic, Safety, Long Context, Math, Multimodal, Multilingual or Instruction Following.
Price
$0.16input$0.64outputper million tokens
From Alibaba's own price page · 5 providers tracked · All prices
Evidence
57results on45benchmarks
- 6 independently verified
- 23 aggregator
- 5 vendor-reported
- 23 cross-referenced
From 7 sources · latest Oct 8, 2026 · How verification works
API features
Tool callingStructured outputsReasoning
As listed by OpenRouter
Research
58 papers reference Qwen3 32BQwen3 32B benchmark results
57 results on 45 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
36.0% behind the leader3 of 4 ranked benchmarks measured
- 5.90May 2, 2026
- AA-Omniscience · Accuracy17.43Oct 8, 2026omniscienceAccuracy
- AA-Omniscience · Non-hallucination17.88Oct 8, 2026omniscienceNonHallucination
0 of 6 ranked benchmarks measured
- 0.29Oct 8, 2026
- GPQA Diamond53.54Oct 8, 2026gpqa
- GPQA Diamond66.77Oct 8, 2026gpqa
Show 6 more reasoning resultsHide 6 reasoning results
- ARC-AGI-20.00Jun 15, 2026ArcAGI V2
- GPQA Diamond66.70Jun 15, 2026GPQA-D
- 68.40Jun 4, 2026
- Humanity's Last Exam4.13Oct 8, 2026aa_hle
- Humanity's Last Exam7.41Oct 8, 2026aa_hle
- 6.90Jun 15, 2026
0 of 10 ranked benchmarks measured
- SciCode36.00Oct 8, 2026aa_scicode
- Terminal-Bench 2.15.24Oct 8, 2026terminalbenchV21
- Terminal-Bench Hard3.03Oct 8, 2026aa_terminalbench_hard
Show 4 more coding resultsHide 4 coding results
- LiveCodeBench v653.40Jun 15, 2026LiveCodeBench v6 (02/2025-05/2025)
- SciCode28.01Sep 4, 2026aa_scicode
- SWE-bench Verified39.70Jun 15, 2026SWE-Bench Verified (AgentLess 4*10)
- SWE-bench Verified23.40Jun 15, 2026SWE-Bench Verified (OpenHands)
0 of 7 ranked benchmarks measured
- 0.00Oct 8, 2026
- Terminal-Bench 4.00.00Oct 8, 2026
- τ-Bench V3 · Banking5.36Oct 8, 2026tauBanking
0 of 3 ranked benchmarks measured
- 0.00Oct 8, 2026
0 of 3 ranked benchmarks measured
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- vectara_avg_summary_length115.80May 2, 2026Average Summary Length (Words)
- vectara_answer_rate99.90May 2, 2026Answer Rate
- vectara_factual_consistency94.10May 2, 2026Factual Consistency Rate
- 40.00May 1, 2026
- 59.90May 26, 2025
- τ²-Bench Telecom (AA run)29.82Oct 8, 2026aa_tau2
Show 26 more resultsHide 26 results
- 0.89Sep 9, 2026
- AA Intelligence7.34Oct 8, 2026aa_intelligence_index
- AA Intelligence8.59Oct 8, 2026aa_intelligence_index
- -50.37Oct 8, 2026
- AIME 202482.70Jun 15, 2026AIME24
- AIME 202481.40Jun 4, 2026AIME 24
- 72.90Oct 7, 2026
- AIME 202573.30Jun 15, 2026AIME25
- AIME 202572.90Jun 4, 2026AIME 25
- 93.80Sep 8, 2026
- Artificial Analysis Coding Index15.30Sep 9, 2026aa_coding_index
- 29.00Jun 15, 2026
- BFCLv470.30Oct 7, 2026BFCL
- 88.40Jun 15, 2026
- 65.40Jun 15, 2026
- 74.90Oct 7, 2026
- 65.70Aug 23, 2026
- 86.20Jun 15, 2026
- 81.80Jun 15, 2026
- 79.00Jun 15, 2026
- 7.70Jun 15, 2026
- RULER77.50Jun 15, 2026RULER (128K)
- 8.60Jun 15, 2026
- 49.30Jun 15, 2026
- TAU-bench (airline)38.00Jun 15, 2026TAU1-Airline
- TAU-bench (retail)40.90Jun 15, 2026TAU1-Retail
Qwen3 32B: common questions
Who makes Qwen3 32B?
Qwen3 32B is made by Alibaba.
When was Qwen3 32B released?
Qwen3 32B was released on Apr 28, 2025, according to Artificial Analysis.
What is Qwen3 32B good at?
Qwen3 32B is behind the leaders in factuality. Too few results yet to rate reasoning, coding, agentic tasks, safety, long context, math, multimodal tasks, multilingual tasks, or instruction following.
How much does Qwen3 32B cost?
Qwen3 32B costs $0.16 per million input tokens and $0.64 per million output tokens, according to Alibaba's own price page. We track its price at 5 providers. At a mix of three input tokens to one output token, it is cheaper than 73% of the 330 priced models we track.
How many benchmarks has Qwen3 32B been tested on?
We track 57 results for Qwen3 32B on 45 benchmarks from 7 sources, 6 of them independently verified. The latest was recorded on Oct 8, 2026.
Which API features does Qwen3 32B support?
OpenRouter lists tool calling, structured outputs, and reasoning for Qwen3 32B.
About this record
Where Qwen3 32B's numbers come from, and every name it appears under.
- Tracked since
- Apr 27, 2026
- Newest source mention
- Jun 1, 2026
Where the results come from
Verification: 57 scores · 6 independently verified · 23 aggregator-attributed · 23 vendor cross-reference · 5 vendor-reported. How these tiers are assigned
From 7 sources on 6 sites. Artificial Analysis supplies 23 of them; the 6 independently verified results come from 3 sites. Bars are coloured by trust tier.
- artificialanalysis.ai23
- huggingface.co23
- api.llm-stats.com5
- raw.githubusercontent.com4
- aider.chat1
- arxiv.org1