---
id: "qwen/qwen3-30b-a3b-thinking-2507"
also: ["qwen/qwen3-30b-a3b-thinking-2507", "Qwen/Qwen3-30B-A3B-Thinking-2507"]
cheapest_list_usd_per_m: "0.2 in, 2.4 out, at alibaba on openrouter"
cheapest_effective_usd_per_m: "0.1998 in, 2.3998 out, at alibaba on openrouter"
context: 81920
max_output: 32768
input: ["text"]
output: ["text"]
released: "2025-08-28"
params: ["frequency_penalty", "include_reasoning", "max_tokens", "presence_penalty", "reasoning", "response_format", "seed", "stop", "temperature", "tool_choice", "tools", "top_k", "top_p"]
endpoints: 1
gateways: "openrouter (1)"
no_training: 1
trains: 0
retains_prompts: 1
intelligence: 9.8000
coding: 12.1000
agentic: 0.9000
---
# Qwen: Qwen3 30B A3B Thinking 2507

Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking. The model is designed specifically for “thinking mode,” where internal reasoning traces are separated...