Alibaba (Qwen) · Released Jul 25, 2025Qwen3 235B-A22B Thinking 2507
Updated large open Qwen3 MoE thinking model for reasoning, math, coding, and tool use.
- Capability
- 143.8ECI · #93 of 165
- Input
- $0.30per 1M tokens
- Output
- $2.90per 1M tokens
- Context
- 262K82K max output
What it costs
- Input
- $0.30per million tokens
- Output
- $2.90per million tokens
- Blended (3:1)
- $0.95pricier than 56% of priced models
Typical monthly bills
| Monthly usage | Estimated cost |
|---|---|
| Side project2M input + 0.5M output tokens | $2.05 |
| Team assistant25M input + 5M output tokens | $22.00 |
| Production app250M input + 50M output tokens | $220.00 |
Median of 11 paid API providers on models.dev.
Independent benchmarks
#93 of 165 scored models
The shaded band is Epoch AI’s confidence range (141.9–145.2); the tick marks the median scored model.
Scores from Epoch AI, run independently of Alibaba (Qwen).
The details
- Lab
- Alibaba (Qwen)
- Released
- Jul 25, 2025
- Context window
- 262,144 tokens
- Max output
- 81,920 tokens
- Inputs
- Text
- Output
- Text
- Reasoning
- Always on
- Tool calling
- Yes
- Structured output
- No
- Weights
- Open weights · Apache 2.0
- Availability
- 12 API providerslisted on models.dev
More from Alibaba (Qwen)
About Qwen3 235B-A22B Thinking 2507
How much does Qwen3 235B-A22B Thinking 2507 cost?
Qwen3 235B-A22B Thinking 2507 costs $0.30 input / $2.90 output per million tokens (median across 11 API providers). At a 3:1 input-to-output mix that is $0.95 per million tokens, more expensive than 56% of the 445 priced models we track.
What is the context window of Qwen3 235B-A22B Thinking 2507?
Qwen3 235B-A22B Thinking 2507 accepts up to 262,144 tokens per request and can write up to 81,920 tokens in one response.
How good is Qwen3 235B-A22B Thinking 2507?
Epoch AI gives Qwen3 235B-A22B Thinking 2507 a Capabilities Index score of 143.8 (likely range 141.9–145.2), ranking it #93 of 165 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 80.1%, OTIS Mock AIME 2024–2025 86.7%, SimpleQA Verified 40.4%.
Is Qwen3 235B-A22B Thinking 2507 open source?
Yes. Alibaba (Qwen) publishes the weights under the Apache 2.0 licence, so it can be downloaded and self-hosted.
What inputs does Qwen3 235B-A22B Thinking 2507 support?
Qwen3 235B-A22B Thinking 2507 accepts text and replies in text. It is a reasoning model, supports tool calling.
When was Qwen3 235B-A22B Thinking 2507 released?
Alibaba (Qwen) released Qwen3 235B-A22B Thinking 2507 on Jul 25, 2025.
What are the best alternatives to Qwen3 235B-A22B Thinking 2507?
The closest current models from other labs on capability, price and release date are GLM-4.7 (Z.ai (Zhipu), ECI 143.7, $0.60 / $2.20), GPT-5 Mini (OpenAI, ECI 145.4, $0.25 / $2.00), DeepSeek R1 0528 (DeepSeek, ECI 141.2, $0.512 / $2.19) and Gemini 2.5 Flash (Google, ECI 140.6, $0.30 / $2.50).