Qwen3-235B-A22B-Thinking-2507
alibaba/qwen3-235b-a22b-thinking-2507By Alibaba (Qwen) · family: qwen · released 2025-07-25 · knowledge: 2025-04
Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.
Capabilities
Model fit scores
0–100 · higher is betterThese scores reward declared capabilities, context size, price and provider availability — they are not benchmark results. Use them as a directional signal alongside your own evaluation.
Coding68
- Tool calling40/40
- Structured output0/20
- Reasoning10/10
- Context window (100K → 1M)8/20
- Provider availability10/10
Agents75
- Tool calling35/35
- Structured output0/25
- Reasoning15/15
- Output token limit15/15
- Provider availability10/10
JSON / structured output50
- Structured output / JSON mode0/50
- Tool calling20/20
- Temperature control10/10
- Price-friendly for high-volume20/20
Cost efficiency79
- Headline price (log-scaled)79/95
- Has prompt-cache pricing0/5
Long context61
- Context window (100K → 2M)51/90
- Has published price for full window10/10
Production-readiness96
- Number of independent providers40/40
- Has published per-token price20/20
- Context window ≥ 8K15/15
- No data inconsistencies across providers6/10
- Official model (not derivative)15/15
Cost Efficiency Index
Open full calculator →Estimated cost using the recommended provider's headline rate. Each scenario fixes average input/output tokens — the assumptions are shown in the third column.
| Scenario | Cost | Assumption |
|---|---|---|
RAG answer per 1,000 RAG answers | $0.55 < $0.01 per request | 5K input tokens (query + 4 retrieved chunks of ~1K each) and a 500-token answer. Typical SaaS knowledge-base bot. |
Support ticket triage per 10,000 tickets | $1.10 < $0.01 per request | 1K input tokens (ticket body + system prompt) and a 100-token JSON classification reply. High-volume customer support. |
Data extraction per 1,000 documents | $0.25 < $0.01 per request | 2K input tokens (a single document page) and a 500-token JSON extraction. ETL / invoice / form pipelines. |
Code review per 1,000 PRs | $0.90 < $0.01 per request | 8K input tokens (diff + surrounding files) and a 1K-token review comment. PR-bot workloads. |
Agent step per 1,000 steps | $1.26 < $0.01 per request | 12K input tokens (long-running tool history) and a 600-token tool-call decision. Cost per agent step. |
Pricing detail
Recommended pricing from wandb · Qwen/Qwen3-235B-A22B-Thinking-2507
Cheapest provider: modelscope · Unknown input + Unknown output
Available on 17 providers
| Provider | Provider model id | Input / 1M | Output / 1M | Context | Released |
|---|---|---|---|---|---|
| ModelScope modelscope | Qwen/Qwen3-235B-A22B-Thinking-2507 | Unknown | Unknown | 262K | 2025-07-25 |
| OpenRouter openrouter | qwen/qwen3-235b-a22b-thinking-2507 | $0.078 | $0.312 | 262K | 2025-07-25 |
| Hugging Face huggingface | Qwen/Qwen3-235B-A22B-Thinking-2507 | $0.300 | $3.00 | 262K | 2025-07-25 |
| SiliconFlow (China) siliconflow-cn | Qwen/Qwen3-235B-A22B-Thinking-2507 | $0.130 | $0.600 | 262K | 2025-07-28 |
| submodel submodel | Qwen/Qwen3-235B-A22B-Thinking-2507 | $0.200 | $0.600 | 262K | 2025-08-23 |
| IO.NET io-net | Qwen/Qwen3-235B-A22B-Thinking-2507 | $0.110 | $0.600 | 262K | 2025-07-01 |
| Jiekou.AI jiekou | qwen/qwen3-235b-a22b-thinking-2507 | $0.300 | $3.00 | 131K | 2026-01 |
| iFlow iflowcn | qwen3-235b-a22b-thinking-2507 | Unknown | Unknown | 256K | 2025-07-01 |
| NovitaAI novita-ai | qwen/qwen3-235b-a22b-thinking-2507 | $0.300 | $3.00 | 131K | 2025-07-25 |
| Weights & Biases wandb | Qwen/Qwen3-235B-A22B-Thinking-2507 | $0.100 | $0.100 | 262K | 2025-07-25 |
| Chutes chutes | Qwen/Qwen3-235B-A22B-Thinking-2507 | $0.110 | $0.600 | 262K | 2025-12-29 |
| Qiniu qiniu-ai | qwen3-235b-a22b-thinking-2507 | Unknown | Unknown | 262K | 2025-08-12 |
| Kilo Gateway kilo | qwen/qwen3-235b-a22b-thinking-2507 | $0.110 | $0.600 | 262K | 2025-07-25 |
| Venice AI venice | qwen3-235b-a22b-thinking-2507 | $0.450 | $3.50 | 128K | 2025-04-29 |
| Synthetic synthetic | hf:Qwen/Qwen3-235B-A22B-Thinking-2507 | $0.650 | $3.00 | 256K | 2025-07-25 |
| SiliconFlow siliconflow | Qwen/Qwen3-235B-A22B-Thinking-2507 | $0.130 | $0.600 | 262K | 2025-07-28 |
| LLM Gateway llmgateway | qwen3-235b-a22b-thinking-2507 | $0.800 | $2.40 | 131K | 2025-07-08 |
Data inconsistencies across providers
- context_window varies: 128000, 131072, 256000, 262000, 262144
- release_date varies (span 247d): 2025-04-29, 2025-07-01, 2025-07-08, 2025-07-25, 2025-07-28, 2025-08-12, 2025-08-23, 2025-12-29, 2026-01
Different providers report different values for this model. Quick facts above use the representative provider; consult the table for per-provider truth.
Frequently asked questions
How much does Qwen3-235B-A22B-Thinking-2507 cost?
Qwen3-235B-A22B-Thinking-2507 costs $0.100 per 1M input tokens and $0.100 per 1M output tokens, sourced from wandb. Cache reads, audio tokens and >200K-context tiers (where applicable) are listed in the Pricing detail block above.
What is the context window of Qwen3-235B-A22B-Thinking-2507?
Qwen3-235B-A22B-Thinking-2507 has a context window of 262K tokens, with a max output of 131K tokens per reply. This is the total combined size of prompt + completion.
Does Qwen3-235B-A22B-Thinking-2507 support tool calling?
Yes. Qwen3-235B-A22B-Thinking-2507 supports tool calling (function calling). This makes it suitable for production agent and automation workloads where the model has to invoke external functions reliably.
Does Qwen3-235B-A22B-Thinking-2507 support structured output / JSON mode?
Support for structured output / JSON-schema-constrained decoding is not reported for Qwen3-235B-A22B-Thinking-2507 in our data source. Verify with Alibaba (Qwen)'s official documentation before relying on it in production.
Can Qwen3-235B-A22B-Thinking-2507 accept image input?
No. Qwen3-235B-A22B-Thinking-2507 only accepts text as input. If you need image input, see our /capabilities/vision list for current vision-capable models.
Is Qwen3-235B-A22B-Thinking-2507 open-weight?
Yes. Qwen3-235B-A22B-Thinking-2507's weights are publicly available, so you can self-host or fine-tune. Note that open weights ≠ open source — the training data and code are typically not released.
What are the best alternatives to Qwen3-235B-A22B-Thinking-2507?
If Qwen3-235B-A22B-Thinking-2507 doesn't fit, consider Qwen3.5 397B-A17B, Qwen3 32B, Qwen3 235B A22B Instruct 2507. Each one targets the same use case — see the Related links below for direct head-to-head pages.
Where does this data come from?
All numbers come from the public models.dev API and are normalised into a single canonical model record. We re-pull daily and write any changes (price, context, capability) to the /changelog page.
Explore more
More Alibaba (Qwen) models
- Qwen3.5 397B-A17B$0.60 in / $3.60 out
- Qwen3 32B$0.70 in / $2.80 out
- Qwen3 235B A22B Instruct 2507$0.10 in / $0.10 out
- Qwen3-Coder 480B-A35B Instruct$1.50 in / $7.50 out
- Qwen3-Next 80B-A3B Instruct$0.50 in / $2.00 out
Capability lists this model is in
Last updated:
Data is sourced from models.dev and normalized for comparison. Prices and capabilities may change. Always verify critical production decisions with the provider's official documentation.