Qwen3 235B-A22B Thinking 2507
Updated large open Qwen3 MoE thinking model for reasoning, math, coding, and tool use
- Strongest at
- Science Better than 4% of the past year's models · one benchmark so far
- Price per 1M tokens
- $0.45 / $3.5 input / output · OpenRouter
- Context window
- 262.1K tokens in · 81.9K out
Capability profile
Where it lands among 90 models released in the past year. 50 is the average model.
Where these scores come from
Each capability combines the published benchmarks listed here, rescaled so 50 is the reference average and 15 points is one standard deviation. A hollow marker rests on a single benchmark.
- Science WeirdML v2
Price against capability
The shaded corner holds anything better and cheaper. Prices blend input and output 3 to 1.
27 models from the past year score higher on Science for less: DeepSeek V4.1 Flash 1017, GPT-6 Luna, GLM-5.3-Flash and 24 more.
OpenRouter $0.45 / $3.5
Epoch benchmark boards
Best result on each board it has run. The tick marks the leader.
Qwen's past year
48 release days in the past year, about one every 8 days. The next is expected around Oct 30, 2026. Forecasts ↗
Specifications
Specifications and prices from Models.dev; availability and limits may differ by provider. Benchmarks from Epoch AI (CC BY 4.0), LiveBench and the other sources named above.