Qwen: Qwen3 235B A22B Thinking 2507

Text input Text output
Author's Description

Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144...

Key Specifications
Cost
$$$$$
Context
131K
Parameters
235B
Released
Jul 25, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Seed Tools Top P Response Format Reasoning Temperature Presence Penalty Include Reasoning Tool Choice Max Tokens
Features

This model supports the following features:

Response Format Tools Reasoning
Performance Summary

Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) model optimized for complex reasoning. While demonstrating exceptional reliability with a 100% success rate across all benchmarks, its speed performance tends to be slower, ranking in the 14th percentile. Similarly, its pricing is positioned at premium levels, ranking in the 4th percentile. The model exhibits significant strengths in specialized areas. It achieves perfect accuracy in General Knowledge, making it the most accurate model at its price point and among models of comparable speed. Its Coding performance is outstanding at 98.0% accuracy (99th percentile), and it excels in Reasoning (98.0% accuracy, 87th percentile) and Mathematics (92.9% accuracy, 68th percentile), aligning with its "thinking-only" optimization. Hallucination rates are low at 94.0% accuracy. However, a notable weakness is its Instruction Following capability, which is relatively low at 26.3% accuracy (23rd percentile). Email Classification and Keyword Topic Relevance Classification show solid, though not top-tier, performance at 99.0% and 90.0% accuracy respectively. Ethics performance is moderate at 98.0% accuracy (34th percentile).

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.149
Completion $1.5

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Alibaba
Alibaba | qwen/qwen3-235b-a22b-thinking-2507 131K $0.149 / 1M tokens $1.5 / 1M tokens
Novita
Novita | qwen/qwen3-235b-a22b-thinking-2507 131K $0.149 / 1M tokens $1.5 / 1M tokens
Chutes
Chutes | qwen/qwen3-235b-a22b-thinking-2507 262K $0.149 / 1M tokens $1.5 / 1M tokens
Novita
Novita | qwen/qwen3-235b-a22b-thinking-2507 131K $0.3 / 1M tokens $3 / 1M tokens
DeepInfra
DeepInfra | qwen/qwen3-235b-a22b-thinking-2507 262K $0.23 / 1M tokens $2.3 / 1M tokens
Parasail
Parasail | qwen/qwen3-235b-a22b-thinking-2507 262K $0.149 / 1M tokens $1.5 / 1M tokens
Together
Together | qwen/qwen3-235b-a22b-thinking-2507 262K $0.149 / 1M tokens $1.5 / 1M tokens
Crusoe
Crusoe | qwen/qwen3-235b-a22b-thinking-2507 262K $0.149 / 1M tokens $1.5 / 1M tokens
Cerebras
Cerebras | qwen/qwen3-235b-a22b-thinking-2507 131K $0.149 / 1M tokens $1.5 / 1M tokens
GMICloud
GMICloud | qwen/qwen3-235b-a22b-thinking-2507 131K $0.149 / 1M tokens $1.5 / 1M tokens
SiliconFlow
SiliconFlow | qwen/qwen3-235b-a22b-thinking-2507 262K $0.149 / 1M tokens $1.5 / 1M tokens
Chutes
Chutes | qwen/qwen3-235b-a22b-thinking-2507 262K $0.149 / 1M tokens $1.5 / 1M tokens
Friendli
Friendli | qwen/qwen3-235b-a22b-thinking-2507 262K $0.149 / 1M tokens $1.5 / 1M tokens
AtlasCloud
AtlasCloud | qwen/qwen3-235b-a22b-thinking-2507 128K $0.28 / 1M tokens $2.3 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by qwen