Alibaba Cloud / Qwen Team model product
QwQ 32B is an LLM from Alibaba Cloud / Qwen Team focused on reasoning in mathematics and programming.
Updated Sep 8, 2026. Default version: QwQ-32B
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for QwQ-32B.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkBFCL | Score66.40% | Rank10 | Participants11 | Percentile10.00% | EvidenceC | Evaluated |
| BenchmarkLiveBench | Score73.10% | Rank22 | Participants38 | Percentile43.24% | EvidenceC | Evaluated |
| BenchmarkMATH-500 | Score90.60% | Rank25 | Participants32 | Percentile22.58% | EvidenceC | Evaluated |
| BenchmarkAIME 2024 | Score79.50% | Rank31 | Participants53 | Percentile42.31% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench | Score63.40% | Rank32 | Participants75 | Percentile58.11% | EvidenceC | Evaluated |
| BenchmarkIFEval | Score83.90% | Rank44 | Participants68 | Percentile35.82% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,328.96 rating | Rank150 | Participants210 | Percentile28.71% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,335.75 rating | Rank158 | Participants210 | Percentile24.88% | EvidenceA | Evaluated |
| BenchmarkGPQA | Score65.20% | Rank160 | Participants247 | Percentile35.37% | EvidenceC | Evaluated |
Preference and agent-evaluation results for the default version.
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categoryjapanese | Rank125 | Rating / score1,263.74 | Votes608 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank125 | Rating / score1,351.90 | Votes397 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank126 | Rating / score1,278.30 | Votes407 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank129 | Rating / score1,379.22 | Votes1,530 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank129 | Rating / score1,312.94 | Votes668 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank130 | Rating / score1,279.43 | Votes407 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank133 | Rating / score1,337.52 | Votes315 | ObservationsN/A | Result date |
| Arenatext style control | Categoryspanish | Rank133 | Rating / score1,354.31 | Votes397 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank134 | Rating / score1,363.15 | Votes1,704 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank135 | Rating / score1,358.46 | Votes1,704 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank135 | Rating / score1,287.46 | Votes1,243 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank135 | Rating / score1,381.54 | Votes1,530 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank136 | Rating / score1,365.17 | Votes1,561 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank137 | Rating / score1,356.09 | Votes1,561 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank138 | Rating / score1,253.22 | Votes608 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank141 | Rating / score1,344.89 | Votes315 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank142 | Rating / score1,347.50 | Votes4,508 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexpert | Rank142 | Rating / score1,360.81 | Votes1,209 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank143 | Rating / score1,319.53 | Votes3,468 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry medicine and healthcare | Rank145 | Rating / score1,339.11 | Votes1,348 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank146 | Rating / score1,322.98 | Votes1,209 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts english | Rank146 | Rating / score1,373.44 | Votes4,633 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank146 | Rating / score1,288.88 | Votes1,243 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank147 | Rating / score1,350.63 | Votes12,889 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank147 | Rating / score1,340.99 | Votes4,633 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank147 | Rating / score1,305.88 | Votes668 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank148 | Rating / score1,336.29 | Votes1,621 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank149 | Rating / score1,347.10 | Votes7,329 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry writing and literature and language | Rank149 | Rating / score1,300.91 | Votes6,101 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry software and it services | Rank149 | Rating / score1,377.30 | Votes7,329 | ObservationsN/A | Result date |
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderAlibaba (China) | Provider model IDqwq-32b | Regionglobal | Input / 1M$0.287 | Output / 1M$0.861 | Context131.1K | Updated |
| ProviderAbacus | Provider model IDQwen/QwQ-32B | Regionglobal | Input / 1M$0.40 | Output / 1M$0.40 | Context32.8K | Updated |
| ProviderCloudflare Workers AI | Provider model ID@cf/qwen/qwq-32b | Regionglobal | Input / 1M$0.66 | Output / 1M$1 | Context24K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for QwQ-32B. Runtime does not affect the capability score.
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionQwQ-32B | Released | LLMBoard23.69 | Parameters32.5B | ContextN/A | Max outputN/A | Open weightsYes | LicenseApache 2.0 |
| VersionQwQ-32B-Preview | Released | LLMBoardN/A | Parameters32.5B | Context32.8K | Max output32.8K | Open weightsYes | LicenseApache 2.0 |
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about QwQ 32B and its available data.
QwQ 32B is a language model focused on advancing AI reasoning, with deep introspection and self-questioning abilities. It has some limitations in language mixing and recursive or endless reasoning patterns.
Data as of 2026-09-08.
Common questions about QwQ 32B.
QwQ 32B's default version was released on Mar 5, 2025.
QwQ 32B's official API price is $0.287 per million input tokens and $0.861 per million output tokens via Alibaba (China). The lowest tracked third-party offer starts at $0.40 input and $0.40 output via Abacus.
QwQ 32B was created by Alibaba Cloud / Qwen Team.
A context window is not available for the default version.
Yes. The default version is marked as open weight under Apache 2.0.
3 provider offerings are linked to the default version.