qwen3-235b-a22b-2507
Large Qwen mixture-of-experts model for advanced reasoning, coding, and multilingual tasks.
Value rank
#5 of 23
intelligence per dollar in our catalogue
Context rank
#7 tied
131K token window
Agentic strength
33%
tau2-bench tool-use success
Benchmarks
Independent scores by Artificial Analysis, compared with the strongest models in our catalogue.
Intelligence index
Coding index
Agentic (tau2-bench)
Best for
Where this model earns its keep.
The numbers
Pricing is live from our platform. Prices per 1M tokens, zero data retention on every request.
| Input price | $0.07 |
| Cache read price | $0.02 |
| Output price | $0.46 |
| Context window | 131K tokens |
| Intelligence / coding index | 18.2 / 22.1 |
| Agentic: tau2 / terminal-bench | 33% / 15% |
| GPQA / MMLU-Pro | 75% / 83% |
Or consider
Close alternatives in the catalogue.
deepseek-r1-0528
Intelligence 20.1 · agentic 37%
deepseek-chat-v3.1
Intelligence 21 · agentic 35%
qwen3-vl-235b-a22b-instruct
Intelligence 14.3 · agentic 35%
Quick start
OpenAI-compatible. Switch in one line.
client = OpenAI(base_url="https://api.tensorx.ai/v1", api_key="tsx-...")
r = client.chat.completions.create(
model="qwen/qwen3-235b-a22b-2507",
messages=[{"role": "user", "content": "Hello"}],
)
Benchmark data from the Artificial Analysis Intelligence Index v4.1, measured independently. Pricing live from the TensorX platform. All inference on EU-sovereign infrastructure with zero data retention.