Qwen3-Next-80B-A3B-Thinking (Reasoning)
by Alibaba Cloud / Qwen Team
Qwen3-Next-80B-A3B-Thinking (Reasoning) is a reasoning model from Alibaba Cloud / Qwen Team, released on September 10, 2025. It has a 66K-token context window and can generate up to 66K output tokens. API pricing is $0.15 per 1M input tokens and $1.5 per 1M output tokens. It ranks #108 of 220 models by Intelligence Score with a score of 23.8.
Qwen3-Next-80B-A3B-Thinking is the thinking variant of the Qwen3-Next series, featuring the same groundbreaking architecture as the instruct model. Leveraging GSPO, it addresses stability and efficiency challenges of hybrid attention + high-sparsity MoE in RL training. It uses Hybrid Attention combining Gated DeltaNet and Gated Attention for efficient ultra-long context modeling, High-Sparsity MoE with 512 experts (10 activated + 1 shared), and Multi-Token Prediction. With 80B total parameters and only 3B activated, it demonstrates outstanding performance on complex reasoning tasks — outperforming Qwen3-30B-A3B-Thinking-2507, Qwen3-32B-Thinking, and even the proprietary Gemini-2.5-Flash-Thinking across multiple benchmarks. Architecture: 48 layers, 15T training tokens, hybrid layout of 12*(3*(Gated DeltaNet->MoE)->(Gated Attention->MoE)). Supports only thinking mode with automatic <think> tag inclusion, may generate longer thinking content.
Key specs
| Provider | Alibaba Cloud / Qwen Team |
|---|---|
| Release date | 2025-09-10 |
| License | open source |
| Context window | 66K tokens |
| Max output | 66K tokens |
| Input price | $0.15 / 1M tokens |
| Output price | $1.5 / 1M tokens |
| Intelligence Score | 23.8 |
Quick answers
What is the context window of Qwen3-Next-80B-A3B-Thinking (Reasoning)?
Qwen3-Next-80B-A3B-Thinking (Reasoning) has a context window of 66K tokens, with a maximum output of 66K tokens per response.
How much does Qwen3-Next-80B-A3B-Thinking (Reasoning) cost?
Qwen3-Next-80B-A3B-Thinking (Reasoning) costs $0.15 per 1M input tokens and $1.5 per 1M output tokens through the API.
Is Qwen3-Next-80B-A3B-Thinking (Reasoning) open source?
Yes. Qwen3-Next-80B-A3B-Thinking (Reasoning) is an open model released under the open source license.
When was Qwen3-Next-80B-A3B-Thinking (Reasoning) released?
Alibaba Cloud / Qwen Team released Qwen3-Next-80B-A3B-Thinking (Reasoning) on September 10, 2025.
How does Qwen3-Next-80B-A3B-Thinking (Reasoning) compare to Claude Opus 5.5 (Reasoning)?
Qwen3-Next-80B-A3B-Thinking (Reasoning) scores 23.8 on the Intelligence Score versus 60.5 for Claude Opus 5.5 (Reasoning), the current #1. Input pricing is $0.15 vs $4 per 1M tokens.
Compare on the AI leaderboard · All AI models
More models from Alibaba Cloud / Qwen Team
- Qwen3.8 Max (Reasoning) — Alibaba Cloud / Qwen Team
- Qwen3.8-Flash-Next (Reasoning) — Alibaba Cloud / Qwen Team
- Qwen3.8 Flash (Reasoning) — Alibaba Cloud / Qwen Team
- Qwen3.8-27B (Reasoning) — Alibaba Cloud / Qwen Team
- Qwen3.7-Plus (Reasoning) — Alibaba Cloud / Qwen Team
- Qwen3.5-397B-A17B (Reasoning) — Alibaba Cloud / Qwen Team