Qwen3 235B-A22B
Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model developed by Qwen, activating 22B parameters per forward pass. It supports seamless switching between a "thinking" mode for complex reasoning, math, and code tasks, and a "non-thinking" mode for general conversational efficiency. The model demonstrates strong reasoning ability, multilingual support (100+ languages and dialects), advanced instruction-following, and agent tool-calling capabilities. It natively handles a 32K token context window and extends up to 131K tokens using YaRN-based scaling.
- Organization
- Alibaba
- Family
- qwen
- Providers
- 6
- Context
- 131,072
- Output limit
- 16,384
- Knowledge
- 2025-04
- Release
- 2025-04
- Updated
- 2025-04
- Weights
- Open
- Input
- text
- Output
- text
- Capabilities
- Tools, Reasoning, Temperature
Providers
Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.
| Provider | Model ID | Context | Output | Input / output · 1M | Reasoning | Tools | Structured | Details |
|---|---|---|---|---|---|---|---|---|
| Hugging Face | Qwen/Qwen3-235B-A22B | 40,960 | 16,384 | $0.20 / $0.80 | Yes | Yes | Yes | View |
| Kilo Gateway | qwen/qwen3-235b-a22b | 131,072 | 8,192 | $0.455 / $1.82 | Yes | Yes | No | View |
| Merge Gateway | qwen/qwen3-235b-a22b | 131,072 | 32,768 | $0.287 / $1.147 | Yes | Yes | Yes | View |
| NanoGPT | qwen/qwen3-235b-a22b | 262,144 | 16,384 | $0.30 / $0.50 | No | Yes | Yes | View |
| Alibabavia OpenRouter | qwen/qwen3-235b-a22b unknown | 131,072 | 8,192 | $0.455 / $1.82 | Yes | Yes | No | View |
| Scaleway | qwen3-235b-a22b-instruct-2507 | 260,000 | 16,384 | $0.75 / $2.25 | Yes | Yes | — | View |
Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.
Pricing
Compare token rates and cache charges across available routes and hosting configurations.
Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.
qwen/qwen3-235b-a22bObserved - Input
- $0.455
- Output
- $1.82
Hosting prices through OpenRouter All endpoints and cache rates
| Endpoint | Input / 1M | Output / 1M | Cache read / 1M | Cache write / 1M |
|---|---|---|---|---|
| Alibabaalibaba | $0.455 | $1.82 | — | — |
Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.
Performance
Review reported latency and output-throughput percentiles from the latest observed rolling window.
Time to first token · milliseconds
All latency percentiles 1 endpoints
| Endpoint | P50 | P75 | P90 | P99 |
|---|---|---|---|---|
| Alibabaalibaba | 429 ms | 662 ms | 864.8 ms | 3,192.68 ms |
Output throughput · tokens per second
All throughput percentiles 1 endpoints
| Endpoint | P50 | P75 | P90 | P99 |
|---|---|---|---|---|
| Alibabaalibaba | 61 t/s | 72 t/s | 76 t/s | 87.16 t/s |
Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.
Uptime
Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.
Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.
Benchmarks
Review attributed evaluation results on each source's original scale.
qwen/qwen3-235b-a22bObserved | Design Arena | Category | Elo | Win rate | Rank |
|---|---|---|---|---|
| models | 3d | 874 | 24.5% | 114 |
| models | codecategories | 1,008 | 38.2% | 114 |
| models | dataviz | 1,004 | 40% | 108 |
| models | gamedev | 940 | 32.9% | 120 |
| models | uicomponent | 966 | 38.6% | 112 |
| models | website | 1,037 | 40.4% | 113 |
Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.
Detailed evaluations
| Evaluation | Configuration | Source | Results | Details |
|---|---|---|---|---|
| models · codecategories | Qwen3-235B-A22Bqwen/qwen3-235b-a22b-04-28 | Design Arena | 1,008 Elo | Details |
| models · 3d | Qwen3-235B-A22Bqwen/qwen3-235b-a22b-04-28 | Design Arena | 874 Elo | Details |
| models · dataviz | Qwen3-235B-A22Bqwen/qwen3-235b-a22b-04-28 | Design Arena | 1,004 Elo | Details |
| models · gamedev | Qwen3-235B-A22Bqwen/qwen3-235b-a22b-04-28 | Design Arena | 940 Elo | Details |
| models · website | Qwen3-235B-A22Bqwen/qwen3-235b-a22b-04-28 | Design Arena | 1,037 Elo | Details |
| models · uicomponent | Qwen3-235B-A22Bqwen/qwen3-235b-a22b-04-28 | Design Arena | 966 Elo | Details |
Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 4, 2026.
Technical details & API
Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.
qwen/qwen3-235b-a22bObserved - Context
- 131,072
- Maximum output
- 8,192
- Tokenizer
- Qwen3
- Input
- text
- Output
- text
- Moderation
- No
- Instruction format
- qwen3
- Knowledge cutoff
- 2025-03-31
- Retirement
- 2026-10-09
- Added to OpenRouter
- 2025-04-28
https://openrouter.ai/api/v1qwen/qwen3-235b-a22bqwen/qwen3-235b-a22b-04-28Qwen/Qwen3-235B-A22BReasoning
- Required
- No
- Enabled by default
- Not reported
- Default effort
- —
- Supported efforts
- —
Supported parameters
frequency_penaltyinclude_reasoningmax_tokenspresence_penaltyreasoningresponse_formatseedstoptemperaturetool_choicetoolstop_ktop_p
Frequently asked questions
Answers based on the model metadata and serving offers currently in the catalog.
How much does Qwen3 235B-A22B cost?
Listed rates start at $0.20 per million input tokens and $0.50 per million output tokens. Prices vary by provider and configuration; see Pricing for details.
Which providers offer Qwen3 235B-A22B?
There are 6 cataloged offers across 6 providers. Compare model IDs, prices and limits in Providers.
What is the context window?
The cataloged model context is 131,072 tokens. Each serving endpoint may apply a different limit.
Does it support tools and structured output?
Tool calling: Yes. Structured output: Not reported. Support can vary by endpoint.
Resources
Catalog history · 2 recorded revisions
Metadata observations since this model was first cataloged. These are not model release versions.
- ·
ae149fce1af9 - ·
b8c225f5d1ca