Model Library
AlibabaReleased July 25, 2025Qwen3 235B A22B 2507
Alibaba · Released July 25, 2025
Intelligence score
20
At a glance
- Intelligence
- 20
- Speed
- 61.4tokens/sec
- Latency
- 2.92stime to first token
- Input price
- $0.70per 1M tokens
- Output price
- $8.40per 1M tokens
- Context window
- 260Ktokens
- Parameters
- 235billions
Summary
Comparison summary
Qwen3 235B A22B 2507 (Reasoning) scores 20 on the Artificial Analysis Intelligence Index, placing it below average among other open weight models of similar size (median: 25). Qwen3 235B A22B 2507 (Reasoning) generates output at 61.4 tokens per second (based on Alibaba's API), which is below average compared to other open weight models of similar size (median: 61.7 t/s). Qwen3 235B A22B 2507 (Reasoning) costs $0.70 per 1M input tokens (somewhat higher than average, median: $0.59) and $8.40 per 1M output tokens (at the higher end, median: $2.20), based on Alibaba's API.
Specifications
Technical specifications
| Model name | Qwen3 235B A22B 2507 |
|---|---|
| Provider | Alibaba |
| Release date | July 25, 2025 |
| Intelligence score | 20 |
| Speed | 61.4 tokens/sec |
| Latency (TTFT) | 2.92s |
| Input price | $0.70 / 1M tokens |
| Output price | $8.40 / 1M tokens |
| Context window | 260,000 tokens |
| Parameters | 235B |
| Reasoning | Yes |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | Apache 2 |
| API availability | — |
| Source | Model Intelligence legacy profile: qwen3-235b-a22b-instruct-2507-reasoning |
Pricing
Token prices
Price per 1M tokens
Qwen3 235B A22B 2507, as listed in the catalog record
- Input$0.70
- Output$8.40
| Token type | Price per 1M |
|---|---|
| Input | $0.70 |
| Output | $8.40 |
Model Intelligence legacy profile: qwen3-235b-a22b-instruct-2507-reasoning
FAQ · 21
Frequently asked questions
Qwen3 235B A22B 2507 (Reasoning) was released on July 25, 2025.
Qwen3 235B A22B 2507 (Reasoning) was created by Alibaba.
Qwen3 235B A22B 2507 (Reasoning) scores 20 on the Artificial Analysis Intelligence Index, placing it below average among other open weight models of similar size (median: 25).
Qwen3 235B A22B 2507 (Reasoning) generates output at 61.4 tokens per second (based on Alibaba's API), which is below average compared to other open weight models of similar size (median: 61.7 t/s).
Qwen3 235B A22B 2507 (Reasoning) has a time to first token (TTFT) of 2.92s (based on Alibaba's API), which is at the higher end compared to other open weight models of similar size (median: 2.01s).
Qwen3 235B A22B 2507 (Reasoning) costs $0.70 per 1M input tokens (somewhat higher than average, median: $0.59) and $8.40 per 1M output tokens (at the higher end, median: $2.20), based on Alibaba's API.
Qwen3 235B A22B 2507 (Reasoning) costs $0.70 per 1M input tokens and $8.40 per 1M output tokens (based on Alibaba's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $1.47 per 1M tokens. Pricing may vary by provider.
When evaluated on the Intelligence Index, Qwen3 235B A22B 2507 (Reasoning) generated 66M output tokens, which is better than average compared to other open weight models of similar size (median: 92M).
Yes, Qwen3 235B A22B 2507 (Reasoning) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
Qwen3 235B A22B 2507 (Reasoning) supports text input.
Qwen3 235B A22B 2507 (Reasoning) supports text output.
No, Qwen3 235B A22B 2507 (Reasoning) does not support image input. It can only process text.
No, Qwen3 235B A22B 2507 (Reasoning) is not multimodal. It only supports text input.
Qwen3 235B A22B 2507 (Reasoning) has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Qwen3 235B A22B 2507 (Reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Qwen3 235B A22B 2507 (Reasoning) has 235 billion parameters (22 billion active).
Qwen3 235B A22B 2507 (Reasoning) is a Mixture of Experts (MoE) model with 235 billion total parameters, but only 22 billion active parameters are used during inference.
Qwen3 235B A22B 2507 (Reasoning) is released under the Apache 2.0 license. This license allows commercial use.
Qwen3 235B A22B 2507 (Reasoning) achieves a score of 20 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, Qwen3 235B A22B 2507 (Reasoning) is available via API through 5 providers.
Qwen3 235B A22B 2507 (Reasoning) is available through 5 API providers.