Qwen3-235B-A22B-Thinking-2507

Overview

Qwen3-235B-A22B-Thinking-2507 is a state-of-the-art thinking-enabled Mixture-of-Experts (MoE) model with 235B total parameters (22B activated). It features 94 layers, 128 experts (8 activated), and supports 262K native context length. This version delivers significantly improved reasoning performance, achieving state-of-the-art results among open-source thinking models on logical reasoning, mathematics, science, coding, and academic benchmarks. Key enhancements include markedly better general capabilities (instruction following, tool usage, text generation), enhanced 256K long-context understanding, and increased thinking depth. The model supports only thinking mode with automatic <think> tag inclusion.

Qwen3-235B-A22B-Thinking-2507 was released on July 25, 2025. API access is available through Fireworks, Novita.

Benchmarks

Qwen3-235B-A22B-Thinking-2507 Performance Across Datasets

Scores sourced from the model's scorecard, paper, or official blog posts

llm-stats.com - Tue Dec 16 2025

Pricing

Pricing, performance, and capabilities for Qwen3-235B-A22B-Thinking-2507 across different providers:

Provider	Input ($/M)	Output ($/M)	Max Input	Max Output	Latency (s)	Throughput	Quantization	Input	Output
Fireworks	$0.30	$3.00	262.1K	131.1K	—	—	—	Text Image Audio Video	Text Image Audio Video
Novitafp8	$0.30	$3.00	256.0K	131.1K	—	—	fp8	Text Image Audio Video	Text Image Audio Video

Price Comparison for Qwen3-235B-A22B-Thinking-2507

Price per 1M input tokens (USD), lower is better

llm-stats.com - Tue Dec 16 2025

Example Outputs

API Access