Qwen: Qwen3 30B A3B Thinking 2507

Text input Text output
Author's Description

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...

Key Specifications
Cost
$$$$
Context
262K
Parameters
30B
Released
Aug 28, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Tools Tool Choice Temperature Include Reasoning Reasoning Presence Penalty Max Tokens Structured Outputs Response Format Frequency Penalty Top P
Features

This model supports the following features:

Structured Outputs Reasoning Response Format Tools
Performance Summary

Qwen3-30B-A3B-Thinking-2507, a 30B parameter Mixture-of-Experts model, demonstrates strong performance in complex reasoning tasks. It consistently ranks among the fastest models and offers highly competitive pricing across all benchmarks. The model exhibits exceptional reliability with a 100% success rate, indicating robust operational stability. In terms of performance across categories, Qwen3-30B-A3B-Thinking-2507 excels in Reasoning (93rd percentile) and Ethics (100% accuracy), showcasing its advanced analytical and moral judgment capabilities. It also performs very well in Email Classification (85th percentile), Coding (78th percentile), General Knowledge (76th percentile), and Mathematics (76th percentile). A notable weakness is its Instruction Following, where it scored 0% accuracy, suggesting a significant area for improvement despite its description as having stronger instruction following. Its hallucination rate is moderate at 84% accuracy, indicating some room for improvement in acknowledging uncertainty. Overall, this model is well-suited for advanced research and agentic applications requiring structured, long-context reasoning, particularly where ethical considerations and complex problem-solving are paramount.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.08
Completion $0.4
Input Cache Read $0.08

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Nebius
Nebius | qwen/qwen3-30b-a3b-thinking-2507 262K $0.08 / 1M tokens $0.4 / 1M tokens
Chutes
Chutes | qwen/qwen3-30b-a3b-thinking-2507 262K $0.08 / 1M tokens $0.4 / 1M tokens
Alibaba
Alibaba | qwen/qwen3-30b-a3b-thinking-2507 81K $0.13 / 1M tokens $1.56 / 1M tokens
SiliconFlow
SiliconFlow | qwen/qwen3-30b-a3b-thinking-2507 262K $0.09 / 1M tokens $0.3 / 1M tokens
Chutes
Chutes | qwen/qwen3-30b-a3b-thinking-2507 262K $0.08 / 1M tokens $0.4 / 1M tokens
Cloudflare
Cloudflare | qwen/qwen3-30b-a3b-thinking-2507 32K $0.08 / 1M tokens $0.4 / 1M tokens
AtlasCloud
AtlasCloud | qwen/qwen3-30b-a3b-thinking-2507 131K $0.08 / 1M tokens $0.4 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by qwen