Alibaba Cloud / Qwen Team model product
Qwen3 Next 80B A3B Thinking
Qwen3 Next 80B A3B Thinking is a thinking-only large language model from Alibaba Cloud / Qwen Team with 80B total parameters and 3B activated parameters.
Updated Oct 23, 2026. Default version: Qwen3-Next-80B-A3B-Thinking
LLMBoard Score38.7Qwen3-Next-80B-A3B-Thinking
Coverage80%22 benchmark families
Context window65.5KTokens
Official input price$0.50Alibaba API
Qwen3 Next 80B A3B Thinking Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Qwen3 Next 80B A3B Thinking Benchmark Results
Benchmark scores for Qwen3-Next-80B-A3B-Thinking.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkCFEval | Score2,071.00 points | Rank02 | Participants2 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkLiveBench 20241125 | Score76.60% | Rank03 | Participants15 | Percentile85.71% | EvidenceC | Evaluated |
| BenchmarkBFCL-v3 | Score72.00% | Rank05 | Participants20 | Percentile78.95% | EvidenceC | Evaluated |
| BenchmarkMulti-IF | Score77.80% | Rank07 | Participants25 | Percentile75.00% | EvidenceC | Evaluated |
| BenchmarkOJBench | Score29.70% | Rank07 | Participants9 | Percentile25.00% | EvidenceC | Evaluated |
| BenchmarkWritingBench | Score84.60% | Rank09 | Participants16 | Percentile46.67% | EvidenceC | Evaluated |
| BenchmarkPolyMATH | Score56.30% | Rank10 | Participants23 | Percentile59.09% | EvidenceC | Evaluated |
| BenchmarkTAU-bench Retail | Score69.60% | Rank11 | Participants25 | Percentile58.33% | EvidenceC | Evaluated |
| BenchmarkArena-Hard v2 | Score62.30% | Rank12 | Participants19 | Percentile38.89% | EvidenceC | Evaluated |
| BenchmarkInclude | Score78.90% | Rank14 | Participants31 | Percentile56.67% | EvidenceC | Evaluated |
| BenchmarkMMLU-ProX | Score78.70% | Rank14 | Participants33 | Percentile59.38% | EvidenceC | Evaluated |
| BenchmarkTau2 Airline | Score60.50% | Rank14 | Participants26 | Percentile48.00% | EvidenceC | Evaluated |
| BenchmarkTAU-bench Airline | Score49.00% | Rank15 | Participants23 | Percentile36.36% | EvidenceC | Evaluated |
| BenchmarkHMMT25 | Score73.90% | Rank18 | Participants28 | Percentile37.04% | EvidenceC | Evaluated |
| BenchmarkMMLU-Redux | Score92.50% | Rank19 | Participants50 | Percentile63.27% | EvidenceC | Evaluated |
| BenchmarkSuperGPQA | Score60.80% | Rank19 | Participants37 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkTau2 Retail | Score67.80% | Rank24 | Participants28 | Percentile14.81% | EvidenceC | Evaluated |
| BenchmarkIFEval | Score88.90% | Rank25 | Participants70 | Percentile65.22% | EvidenceC | Evaluated |
| BenchmarkTau2 Telecom | Score43.90% | Rank34 | Participants37 | Percentile8.33% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score82.70% | Rank39 | Participants142 | Percentile73.05% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench v6 | Score68.70% | Rank45 | Participants67 | Percentile33.33% | EvidenceC | Evaluated |
| BenchmarkAIME 2025 | Score87.80% | Rank56 | Participants122 | Percentile54.55% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score77.20% | Rank116 | Participants256 | Percentile54.90% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,368.16 rating | Rank133 | Participants218 | Percentile39.17% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,369.26 rating | Rank141 | Participants218 | Percentile35.48% | EvidenceA | Evaluated |
Qwen3 Next 80B A3B Thinking Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categorymath | Rank117 | Rating / score1,398.67 | Votes822 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry medicine and healthcare | Rank120 | Rating / score1,392.04 | Votes708 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank123 | Rating / score1,355.84 | Votes296 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank124 | Rating / score1,392.25 | Votes5,907 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank124 | Rating / score1,309.97 | Votes356 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank124 | Rating / score1,361.62 | Votes296 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank125 | Rating / score1,384.33 | Votes2,104 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank125 | Rating / score1,344.91 | Votes709 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank127 | Rating / score1,407.00 | Votes1,059 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank127 | Rating / score1,358.49 | Votes709 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank129 | Rating / score1,359.93 | Votes2,489 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank129 | Rating / score1,392.50 | Votes822 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank130 | Rating / score1,390.08 | Votes687 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank130 | Rating / score1,393.45 | Votes4,739 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank130 | Rating / score1,280.25 | Votes206 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank130 | Rating / score1,320.34 | Votes356 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank131 | Rating / score1,391.52 | Votes2,668 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank131 | Rating / score1,375.60 | Votes661 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank131 | Rating / score1,384.11 | Votes3,018 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank132 | Rating / score1,293.07 | Votes206 | ObservationsN/A | Result date |
| Arenatext | Categoryexclude ties | Rank133 | Rating / score1,334.02 | Votes9,587 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank133 | Rating / score1,368.16 | Votes13,479 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank133 | Rating / score1,357.64 | Votes456 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry medicine and healthcare | Rank134 | Rating / score1,395.59 | Votes708 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank136 | Rating / score1,351.43 | Votes218 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank136 | Rating / score1,371.29 | Votes6,666 | ObservationsN/A | Result date |
| Arenatext | Categorynon english | Rank137 | Rating / score1,342.49 | Votes7,572 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank137 | Rating / score1,402.51 | Votes1,059 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank138 | Rating / score1,363.51 | Votes841 | ObservationsN/A | Result date |
| Arenatext | Categorylonger query | Rank138 | Rating / score1,352.95 | Votes2,884 | ObservationsN/A | Result date |
Qwen3 Next 80B A3B Thinking Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.50 input, $6 output per 1M
- Official provider
- Alibaba
- Lowest third-party
- From $0.1449 input, $1.45 output per 1M via Qiniu
- Tracked offerings
- 11
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderAlibaba (China) | Provider model IDqwen3-next-80b-a3b-thinking | Regionglobal | Input / 1M$0.144 | Output / 1M$1.43 | Context131.1K | Updated |
| ProviderQiniu | Provider model IDqwen3-next-80b-a3b-thinking | Regionglobal | Input / 1M$0.1449 | Output / 1M$1.45 | Context131.1K | Updated |
| ProviderNanoGPT | Provider model IDqwen/qwen3-next-80b-a3b-thinking | Regionglobal | Input / 1M$0.15 | Output / 1M$0.65 | Context256K | Updated |
| ProviderJalapeno Cloud | Provider model IDQwen3-Next-80B-A3B-Thinking | Regionglobal | Input / 1M$0.15 | Output / 1M$1.5 | Context262.1K | Updated |
| ProviderJiekou.AI | Provider model IDqwen/qwen3-next-80b-a3b-thinking | Regionglobal | Input / 1M$0.15 | Output / 1M$1.5 | Context65.5K | Updated |
| ProviderOpenRouter | Provider model IDqwen/qwen3-next-80b-a3b-thinking | Regionglobal | Input / 1M$0.15 | Output / 1M$1.2 | Context262.1K | Updated |
| ProviderMerge Gateway | Provider model IDqwen/qwen3-next-80b-a3b-thinking | Regionglobal | Input / 1M$0.15 | Output / 1M$1.2 | Context131.1K | Updated |
| ProviderVercel AI Gateway | Provider model IDalibaba/qwen3-next-80b-a3b-thinking | Regionglobal | Input / 1M$0.15 | Output / 1M$1.2 | Context262.1K | Updated |
| ProviderKilo Gateway | Provider model IDqwen/qwen3-next-80b-a3b-thinking | Regionglobal | Input / 1M$0.15 | Output / 1M$1.2 | Context262.1K | Updated |
| ProviderHugging Face | Provider model IDQwen/Qwen3-Next-80B-A3B-Thinking | Regionglobal | Input / 1M$0.30 | Output / 1M$2 | Context262.1K | Updated |
| ProviderAlibaba | Provider model IDqwen3-next-80b-a3b-thinking | Regionglobal | Input / 1M$0.50 | Output / 1M$6 | Context131.1K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Qwen3 Next 80B A3B Thinking Runtime Performance
Provider-specific output speed and catalog latency for Qwen3-Next-80B-A3B-Thinking. Runtime does not affect the capability score.
No runtime data
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.
Qwen3 Next 80B A3B Thinking Specifications
Technical details for the model's default version.
- Version
- Qwen3-Next-80B-A3B-Thinking
- Released
- Sep 10, 2025
- Knowledge cutoff
- Unknown
- Parameters
- 80B
- Context window
- 65.5K
- Max output
- 65.5K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- Apache 2.0
Qwen3 Next 80B A3B Thinking Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionQwen3-Next-80B-A3B-Thinking | Released | LLMBoard38.68 | Parameters80B | Context65.5K | Max output65.5K | Open weightsYes | LicenseApache 2.0 |
Models similar to Qwen3 Next 80B A3B Thinking
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AC
Qwen3 VL 32B Thinking
Alibaba Cloud / Qwen Team
AC
Qwen3.5 9B
Alibaba Cloud / Qwen Team
AC
Qwen3 235B A22B
Alibaba Cloud / Qwen Team
AC
Qwen3 VL 235B A22B
Alibaba Cloud / Qwen Team
AC
Qwen3 Max
Alibaba Cloud / Qwen Team
AC
Qwen3 Next 80B A3B
Alibaba Cloud / Qwen Team
What is Qwen3 Next 80B A3B Thinking?
Key information about Qwen3 Next 80B A3B Thinking and its available data.
Qwen3 Next 80B A3B Thinking is the thinking variant of the Qwen3-Next series, using hybrid attention, a high-sparsity mixture-of-experts architecture with 512 experts, and Multi-Token Prediction. It supports only thinking mode with automatic <think> tag inclusion.
Data as of 2026-10-22.
FAQ
Common questions about Qwen3 Next 80B A3B Thinking.
When was Qwen3 Next 80B A3B Thinking released?
Qwen3 Next 80B A3B Thinking's default version was released on Sep 10, 2025.
How much does Qwen3 Next 80B A3B Thinking cost?
Qwen3 Next 80B A3B Thinking's official API price is $0.50 per million input tokens and $6 per million output tokens via Alibaba. The lowest tracked third-party offer starts at $0.1449 input and $1.45 output via Qiniu.
Who created Qwen3 Next 80B A3B Thinking?
Qwen3 Next 80B A3B Thinking was created by Alibaba Cloud / Qwen Team.
What is the context window for Qwen3 Next 80B A3B Thinking?
The default version has a 65.5K token context window.
Is Qwen3 Next 80B A3B Thinking open weight?
Yes. The default version is marked as open weight under Apache 2.0.