Qwen3-30B-A3B-Instruct-2507
Qwen3-30B-A3B-Instruct-2507 is an AI model published by 阿里巴巴, released on 2025-07-29, for 聊天大模型, with 305.0B parameters, and 256K tokens context length, 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
| Type | Condition | Input | Output |
|---|---|---|---|
| Text | - | $0.500/ 1M | $3.00/ 1M |
| Type | TTL | Write | Read |
|---|---|---|---|
| Text | - | $0.625/ 1M | $0.050/ 1M |
Qwen3-30B-A3B-2507 currently shows benchmark results led by Creative Writing (7 / 22, score 86), MMLU Pro (58 / 117, score 78.40), GPQA Diamond (100 / 166, score 70.40). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.
| Variant name | Version type | Quantization | Model size | HuggingFace link |
|---|---|---|---|---|
| Qwen3-30B-A3B-Thinking-2507ℹ️ | Instruct | FP16 | 61.1 GB | Download link |
| Qwen3-30B-A3B-Instruct-2507-FP8ℹ️ | Instruct | FP8 | 31.17GB | Download link |
Qwen3-30B-A3B-Instruct-2507是阿里开源的Qwen3-30B-A3B模型升级版本，前者是2025年4月28日阿里开源的混合思考模式的模型（即一个模型支持推理和非推理模式），三个月后阿里发布了新版本，将其拆分为只支持推理和不支持推理的两个版本。Qwen3-30B-A3B-Instruct-2507是其中仅支持非推理模式的版本，但是效果提升明显。
Qwen3-30B-A3B-Instruct-2507模型与其它模型的评测结果对比：
欢迎关注 DataLearner 官方微信，获得最新 AI 技术推送
