Qwen3 Next 80B A3B
Qwen3 Next 80B A3B by Alibaba - a sparse mixture-of-experts model with hybrid attention and high-sparsity MoE for efficient long-context inference.
Model Overview
Capabilities, design details, and architectural traits
Qwen3 Next 80B A3B - sparse mixture-of-experts model
Qwen3 Next 80B A3B is the Mixture-of-Experts member of the Qwen3-Next family designed to activate a small subset of experts per inference for efficiency. The model emphasizes sparse activation and hybrid attention to deliver long-context reasoning with lower active compute per query.
Key technical focus
- High-sparsity MoE lets the model keep a large parameter capacity while activating only a small subset of experts per request for improved throughput.
- Hybrid attention / gated mechanisms are used to maintain stability and scaling efficiency on ultra-long contexts.
Intended usage
The model is intended for production scenarios that need large-model reasoning, especially tasks requiring sustained long-context understanding and retrieval-augmented workflows.
Benchmark Performance
Independent evaluations · Artificial Analysis
Accuracy & Capability Details
Compare Models Side-by-Side
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | Alibaba | Anthropic | Anthropic |
| Release Date | September 11, 2025 | September 22, 2026 | September 28, 2026 |
| Knowledge Cutoff | Sep 2025 | - | Jun 2026 |
| Context & Limits | |||
| Context Window | 262K | 1M Best Context Window | 1M Best Context Window |
| Pricing (per 1M tokens) | |||
| Input Pricing | $0.15 Best Input Pricing | $4 | $2 |
| Output Pricing | $1.10 Best Output Pricing | $20 | $10 |
| Modalities | |||
| Inputs | text | textimagefile | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 9.6 | 57.6 Best Intelligence Index | 56.0 |
| Coding Index | - | - | - |
| Agentic Index | - | - | - |
Humanity's Last Exam
Extremely difficult logical reasoning and knowledge.
Long Context Reasoning
Logical reasoning over long context windows.
SciCode Benchmark
Scientific coding and mathematical modeling.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
Explore more from Alibaba
Other models by Alibaba
Qwen3.8 Max
Qwen3.8 2.4T A95B
Qwen3.8-Flash-Next
Qwen3.8 27B
Top AI Models
Leading alternatives by intelligence score
