Alibaba Cloud / Qwen Team model product
Qwen3 Next 80B A3B
Qwen3-Next-80B-A3B-Instruct is a language model with 80B total parameters and 3B activated parameters, using hybrid attention, a sparse mixture-of-experts architecture, and multi-token prediction.
Updated Sep 8, 2026. Default version: Qwen3-Next-80B-A3B-Instruct
LLMBoard Score35.1Qwen3-Next-80B-A3B-Instruct
Coverage80%24 benchmark families
Context window65.5KTokens
Official input price$0.50Alibaba API
Qwen3 Next 80B A3B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Qwen3 Next 80B A3B Benchmark Results
Benchmark scores for Qwen3-Next-80B-A3B-Instruct.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkArena-Hard v2 | Score82.70% | Rank02 | Participants19 | Percentile94.44% | EvidenceC | Evaluated |
| BenchmarkMultiPL-E | Score87.80% | Rank02 | Participants13 | Percentile91.67% | EvidenceC | Evaluated |
| BenchmarkWritingBench | Score87.30% | Rank02 | Participants15 | Percentile92.86% | EvidenceC | Evaluated |
| BenchmarkLiveBench 20241125 | Score75.80% | Rank04 | Participants14 | Percentile76.92% | EvidenceC | Evaluated |
| BenchmarkCreative Writing v3 | Score85.30% | Rank08 | Participants13 | Percentile41.67% | EvidenceC | Evaluated |
| BenchmarkMulti-IF | Score75.80% | Rank09 | Participants23 | Percentile63.64% | EvidenceC | Evaluated |
| BenchmarkBFCL-v3 | Score70.30% | Rank10 | Participants19 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkInclude | Score78.90% | Rank13 | Participants31 | Percentile60.00% | EvidenceC | Evaluated |
| BenchmarkAider-Polyglot | Score49.80% | Rank16 | Participants22 | Percentile28.57% | EvidenceC | Evaluated |
| BenchmarkMMLU-ProX | Score76.70% | Rank16 | Participants32 | Percentile51.61% | EvidenceC | Evaluated |
| BenchmarkPolyMATH | Score45.90% | Rank16 | Participants23 | Percentile31.82% | EvidenceC | Evaluated |
| BenchmarkTAU-bench Airline | Score44.00% | Rank18 | Participants23 | Percentile22.73% | EvidenceC | Evaluated |
| BenchmarkSuperGPQA | Score58.80% | Rank19 | Participants34 | Percentile45.45% | EvidenceC | Evaluated |
| BenchmarkTAU-bench Retail | Score60.90% | Rank19 | Participants25 | Percentile25.00% | EvidenceC | Evaluated |
| BenchmarkMMLU-Redux | Score90.90% | Rank23 | Participants48 | Percentile53.19% | EvidenceC | Evaluated |
| BenchmarkTau2 Airline | Score45.50% | Rank23 | Participants24 | Percentile4.35% | EvidenceC | Evaluated |
| BenchmarkHMMT25 | Score54.10% | Rank24 | Participants28 | Percentile14.81% | EvidenceC | Evaluated |
| BenchmarkTau2 Retail | Score57.30% | Rank26 | Participants27 | Percentile3.85% | EvidenceC | Evaluated |
| BenchmarkIFEval | Score87.60% | Rank32 | Participants68 | Percentile53.73% | EvidenceC | Evaluated |
| BenchmarkTau2 Telecom | Score13.20% | Rank36 | Participants36 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench v6 | Score56.60% | Rank48 | Participants62 | Percentile22.95% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score80.60% | Rank57 | Participants138 | Percentile59.12% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,417.26 rating | Rank85 | Participants210 | Percentile59.81% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,418.35 rating | Rank91 | Participants121 | Percentile25.00% | EvidenceA | Evaluated |
| BenchmarkAIME 2025 | Score69.50% | Rank94 | Participants119 | Percentile21.19% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,399.21 rating | Rank110 | Participants210 | Percentile47.85% | EvidenceA | Evaluated |
| BenchmarkGPQA | Score72.90% | Rank131 | Participants247 | Percentile47.15% | EvidenceC | Evaluated |
Qwen3 Next 80B A3B Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categoryindustry medicine and healthcare | Rank44 | Rating / score1,456.04 | Votes1,175 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank52 | Rating / score1,441.62 | Votes1,191 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank54 | Rating / score1,398.30 | Votes310 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank54 | Rating / score1,435.28 | Votes684 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank67 | Rating / score1,417.53 | Votes487 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank70 | Rating / score1,433.77 | Votes1,034 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank71 | Rating / score1,420.74 | Votes4,259 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry business and management and financial operations | Rank73 | Rating / score1,431.62 | Votes799 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank74 | Rating / score1,461.25 | Votes1,712 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank74 | Rating / score1,446.79 | Votes7,935 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank74 | Rating / score1,372.91 | Votes310 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank75 | Rating / score1,431.41 | Votes1,399 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank76 | Rating / score1,367.83 | Votes513 | ObservationsN/A | Result date |
| Arenatext | Categorynon english | Rank79 | Rating / score1,406.96 | Votes12,863 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank80 | Rating / score1,440.67 | Votes4,734 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank80 | Rating / score1,427.34 | Votes11,834 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank80 | Rating / score1,433.92 | Votes3,505 | ObservationsN/A | Result date |
| Arenatext factuality | Categorynon english | Rank82 | Rating / score1,412.15 | Votes3,705 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank84 | Rating / score1,397.99 | Votes1,013 | ObservationsN/A | Result date |
| Arenatext factuality | Categorymulti turn | Rank84 | Rating / score1,422.97 | Votes713 | ObservationsN/A | Result date |
| Arenatext | Categoryexclude ties | Rank85 | Rating / score1,404.56 | Votes15,906 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank85 | Rating / score1,431.12 | Votes5,144 | ObservationsN/A | Result date |
| Arenatext | Categorymulti turn | Rank85 | Rating / score1,415.67 | Votes3,987 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank85 | Rating / score1,417.26 | Votes22,532 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry medicine and healthcare | Rank85 | Rating / score1,438.98 | Votes1,175 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryexclude ties | Rank88 | Rating / score1,410.02 | Votes5,177 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank89 | Rating / score1,421.80 | Votes9,636 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank89 | Rating / score1,401.69 | Votes1,567 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryhard prompts | Rank89 | Rating / score1,442.85 | Votes3,746 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank90 | Rating / score1,417.20 | Votes1,191 | ObservationsN/A | Result date |
Qwen3 Next 80B A3B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.50 input, $2 output per 1M
- Official provider
- Alibaba
- Lowest third-party
- From $0.09 input, $1.1 output per 1M via Deep Infra
- Tracked offerings
- 19
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderNvidia | Provider model IDqwen/qwen3-next-80b-a3b-instruct | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context262.1K | Updated |
| ProviderDeep Infra | Provider model IDQwen/Qwen3-Next-80B-A3B-Instruct | Regionglobal | Input / 1M$0.09 | Output / 1M$1.1 | Context262.1K | Updated |
| ProviderKilo Gateway | Provider model IDqwen/qwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.0975 | Output / 1M$0.78 | Context262.1K | Updated |
| ProviderOpenRouter | Provider model IDqwen/qwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.10 | Output / 1M$1.1 | Context262.1K | Updated |
| ProviderCharm Hyper | Provider model IDqwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.1175 | Output / 1M$1.14 | Context262.1K | Updated |
| ProviderHelicone | Provider model IDqwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.14 | Output / 1M$1.4 | Context262K | Updated |
| ProviderAlibaba (China) | Provider model IDqwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.144 | Output / 1M$0.574 | Context131.1K | Updated |
| ProviderMerge Gateway | Provider model IDqwen/qwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.144 | Output / 1M$0.574 | Context131.1K | Updated |
| ProviderNanoGPT | Provider model IDqwen/Qwen3-Next-80B-A3B-Instruct | Regionglobal | Input / 1M$0.15 | Output / 1M$0.65 | Context262.1K | Updated |
| ProviderJalapeno Cloud | Provider model IDQwen3-Next-80B-A3B-Instruct | Regionglobal | Input / 1M$0.15 | Output / 1M$1.5 | Context129K | Updated |
| ProviderJiekou.AI | Provider model IDqwen/qwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.15 | Output / 1M$1.5 | Context65.5K | Updated |
| ProviderNeon | Provider model IDqwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.15 | Output / 1M$1.2 | Context131.1K | Updated |
| ProviderVercel AI Gateway | Provider model IDalibaba/qwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.15 | Output / 1M$1.2 | Context131.1K | Updated |
| ProviderEden AI | Provider model IDqwen/qwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.15 | Output / 1M$1.2 | Context131.1K | Updated |
| ProviderNovitaAI | Provider model IDqwen/qwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.15 | Output / 1M$1.5 | Context131.1K | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDqwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.15 | Output / 1M$1.2 | Context131.1K | Updated |
| ProviderHugging Face | Provider model IDQwen/Qwen3-Next-80B-A3B-Instruct | Regionglobal | Input / 1M$0.25 | Output / 1M$1 | Context262.1K | Updated |
| ProviderAlibaba | Provider model IDqwen3-next-80b-a3b-instruct | Regionglobal | Input / 1M$0.50 | Output / 1M$2 | Context131.1K | Updated |
Qwen3 Next 80B A3B Runtime Performance
Provider-specific output speed and catalog latency for Qwen3-Next-80B-A3B-Instruct. 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 Specifications
Technical details for the model's default version.
- Version
- Qwen3-Next-80B-A3B-Instruct
- 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 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-Base | Released | LLMBoardN/A | Parameters80B | ContextN/A | Max outputN/A | Open weightsYes | LicenseApache 2.0 |
| VersionQwen3-Next-80B-A3B-Instruct | Released | LLMBoard35.15 | Parameters80B | Context65.5K | Max output65.5K | Open weightsYes | LicenseApache 2.0 |
Models similar to Qwen3 Next 80B A3B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AC
Qwen3 Coder 480B A35B
Alibaba Cloud / Qwen Team
AC
Qwen3 VL 32B
Alibaba Cloud / Qwen Team
AC
Qwen3.5 4B
Alibaba Cloud / Qwen Team
AC
Qwen3 30B A3B
Alibaba Cloud / Qwen Team
AC
Qwen3 VL 32B Thinking
Alibaba Cloud / Qwen Team
AC
Qwen3 Next 80B A3B Thinking
Alibaba Cloud / Qwen Team
What is Qwen3 Next 80B A3B?
Key information about Qwen3 Next 80B A3B and its available data.
Qwen3-Next-80B-A3B-Instruct is the first model in the Qwen3-Next series from Alibaba Cloud / Qwen Team. It has 48 layers, was trained on 15T tokens, and is designed for ultra-long-context tasks up to 256K tokens, extensible to 1M with YaRN.
Data as of 2026-09-08.
FAQ
Common questions about Qwen3 Next 80B A3B.
When was Qwen3 Next 80B A3B released?
Qwen3 Next 80B A3B's default version was released on Sep 10, 2025.
How much does Qwen3 Next 80B A3B cost?
Qwen3 Next 80B A3B's official API price is $0.50 per million input tokens and $2 per million output tokens via Alibaba. The lowest tracked third-party offer starts at $0.09 input and $1.1 output via Deep Infra.
Who created Qwen3 Next 80B A3B?
Qwen3 Next 80B A3B was created by Alibaba Cloud / Qwen Team.
What is the context window for Qwen3 Next 80B A3B?
The default version has a 65.5K token context window.
Is Qwen3 Next 80B A3B open weight?
Yes. The default version is marked as open weight under Apache 2.0.