Qwen: Qwen3 235B A22B Thinking 2507

By Qwen

Released: 2025-07-25

Visit Website

API

LLM

RAG

Qwen

Paid

New

Qwen Qwen3 235B A22B Thinking 2507, developed by Alibaba Qwen, features 235B parameters, MoE architecture and 262k-token context window. Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimised for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144 tokens of context. This "thinking-only" variant enhances structured logical reasoning, mathematics, science, and long-form generation, showing strong benchmark performance across AIME, SuperGPQA, LiveCodeBench, and MMLU-Redux. It enforces a special reasoning mode (</think>) and is designed for high-token outputs (up to 81,920 tokens) in challenging domains. The model is instruction-tuned and excels at step-by-step reasoning, tool use, agentic workflows, and multilingual tasks. This release represents the most capable open-source variant in the Qwen3-235B series, surpassing many closed models in structured reasoning use cases. Available at $0.11/1M tokens.

Visit Qwen: Qwen3 235B A22B Thinking 2507

AI-Powered

Leverages advanced AI technology to deliver cutting-edge capabilities and results.

Fast & Efficient

Optimized performance ensures quick results without compromising on quality.

Purpose-Built

Specifically designed for llms tasks and workflows.

Qwen Model Timeline

23 Oct 2025Qwen: Qwen3 VL 32B Instruct

262k tokens context

14 Oct 2025Qwen: Qwen3 VL 8B Thinking

256k tokens context

14 Oct 2025Qwen: Qwen3 VL 8B Instruct

131k tokens context

6 Oct 2025Qwen: Qwen3 VL 30B A3B Thinking

131k tokens context

6 Oct 2025Qwen: Qwen3 VL 30B A3B Instruct

262k tokens context

23 Sept 2025Qwen: Qwen3 VL 235B A22B Thinking

262k tokens context

23 Sept 2025Qwen: Qwen3 VL 235B A22B Instruct

262k tokens context

23 Sept 2025Qwen: Qwen3 Max

256k tokens context

23 Sept 2025Qwen: Qwen3 Coder Plus

128k tokens context

17 Sept 2025Qwen: Qwen3 Coder Flash

128k tokens context

11 Sept 2025Qwen: Qwen3 Next 80B A3B Thinking

128k tokens context

11 Sept 2025Qwen: Qwen3 Next 80B A3B Instruct (free)

262k tokens context

11 Sept 2025Qwen: Qwen3 Next 80B A3B Instruct

262k tokens context

8 Sept 2025Qwen: Qwen Plus 0728

1,000k tokens context

8 Sept 2025Qwen: Qwen Plus 0728 (thinking)

1,000k tokens context

28 Aug 2025Qwen: Qwen3 30B A3B Thinking 2507

33k tokens context

31 Jul 2025Qwen: Qwen3 Coder 30B A3B Instruct

160k tokens context

29 Jul 2025Qwen: Qwen3 30B A3B Instruct 2507

262k tokens context

25 Jul 2025

Qwen: Qwen3 235B A22B Thinking 2507Current

262k tokens context

23 Jul 2025Qwen: Qwen3 Coder 480B A35B (free)

262k tokens context

23 Jul 2025Qwen: Qwen3 Coder 480B A35B

262k tokens context

23 Jul 2025Qwen: Qwen3 Coder 480B A35B (exacto)

262k tokens context

21 Jul 2025Qwen: Qwen3 235B A22B Instruct 2507

262k tokens context

30 Apr 2025Qwen: Qwen3 4B (free)

41k tokens context

28 Apr 2025Qwen: Qwen3 30B A3B

41k tokens context

28 Apr 2025Qwen: Qwen3 8B

32k tokens context

28 Apr 2025Qwen: Qwen3 14B

41k tokens context

28 Apr 2025Qwen: Qwen3 32B

41k tokens context

28 Apr 2025Qwen: Qwen3 235B A22B

41k tokens context

15 Apr 2025Qwen: Qwen2.5 Coder 7B Instruct

33k tokens context

24 Mar 2025Qwen: Qwen2.5 VL 32B Instruct

16k tokens context

5 Mar 2025Qwen: QwQ 32B

33k tokens context

5 Feb 2025Qwen: Qwen VL Plus

8k tokens context

1 Feb 2025Qwen: Qwen VL Max

131k tokens context

1 Feb 2025Qwen: Qwen-Turbo

1,000k tokens context

1 Feb 2025Qwen: Qwen2.5 VL 72B Instruct

33k tokens context

1 Feb 2025Qwen: Qwen-Plus

131k tokens context

1 Feb 2025Qwen: Qwen-Max

33k tokens context

11 Nov 2024Qwen2.5 Coder 32B Instruct

33k tokens context

16 Oct 2024Qwen: Qwen2.5 7B Instruct

33k tokens context

19 Sept 2024Qwen2.5 72B Instruct

33k tokens context

28 Aug 2024Qwen: Qwen2.5-VL 7B Instruct (free)

33k tokens context

28 Aug 2024Qwen: Qwen2.5-VL 7B Instruct

33k tokens context

Specifications

pricing$0.11 / $0.60 (per 1M)

context Window262k tokens

AI Evaluation

4.8

Expert Rating

Text4.9/5

Coding3.5/5

Built for deep analytical thinking and multi-step problem solving. Excels at tasks requiring careful logical reasoning and systematic analysis.

Pros

Competitive pricing ($0.11/1M)
262k token context window
Large-scale 235B architecture
Advanced logical reasoning

Cons

Requires substantial compute
API integration required

Related Tools

FLUX

FLUX, from Black Forest Labs, is a family of high-quality open and commercial image generation models prized for photorealism and prompt adherence. Widely integrated across third-party tools and APIs, it has become a default backbone for image generation.

Claude Opus 4.8

Claude Opus 4.8 is Anthropic's June 2026 flagship model, succeeding Opus 4.7. It posts a headline score of 81 on the hardest agentic coding and reasoning suites, holds long-horizon tool-use plans together across far more steps, and is notably more candid about its own uncertainty - refusing to fabricate rather than confidently pressing on. It is the default choice for serious agentic and software-engineering workloads.

Anthropic Mythos 1

Mythos 1 is the full release of Anthropic's specialised, security-focused system, graduating from the guarded Claude Mythos Preview. Built around Project Glasswing, it pairs frontier reasoning with defensive security tooling - vulnerability analysis, triage and incident response - deployed with heavy oversight and auditability. It runs alongside the general-purpose Claude family rather than replacing it.