Qwen3-30B-A3B
Qwen3-30B-A3B is an AI model published by 阿里巴巴, released on 2025-04-28, for 推理大模型, with 305.0B parameters, and 128K tokens context length, requiring about 61.09 GB storage, under the Apache 2.0 license.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Modality | Input | Output |
|---|---|---|
| Text | $0.2 | $0.8 |
Qwen3-30B-A3B currently shows benchmark results led by MMLU Pro (82 / 116, score 69.10), GPQA Diamond (135 / 166, score 54.80), LiveBench (48 / 51, score 57.80). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.
Qwen3-30B-A3B是阿里开源的第三代千问大模型中的MoE架构版本的模型，参数300亿，每次推理激活其中的30亿参数。
2025年7月29日，阿里发布了这个模型的更新版本，仅支持非推理模式的Qwen3-30B-A3B-Instruct-2507，但是各方面能力提升明显。
欢迎关注 DataLearner 官方微信，获得最新 AI 技术推送
