Alibaba: qwen3-235b-a22b-thinking-2507:free

qwen3-235b-a22b-thinking-2507:free
256K context262.1K outReasoningToolsStructured
Released Jul 25, 2025Knowledge cutoff 2025Updated Jul 25, 2025

The flagship Mixture-of-Experts (MoE) reasoning model released by Alibaba Cloud on July 25, 2025. As the pinnacle of the Qwen3 series, it features a 235B total parameter architecture specifically post-trained for deep Chain-of-Thought (CoT) reasoning. It natively generates <think> blocks to reveal its analytical process before providing a final answer. Boasting a staggering 92.3 on the AIME 2025 benchmark, its reasoning prowess rivals OpenAI o1, o3-mini, and DeepSeek-R1, making it the definitive choice for scientific research, advanced software engineering, and complex strategic planning.

Mode chatTokenizer Qwen3Quantization fp8Qwen/Qwen3-235B-A22B-Thinking-2507

All providers for this model are busy right now

Every upstream provider has hit its rate limit. The model comes back automatically once limits lift, usually within hours. Try again in a little while or switch to another model.

Request this model on Discord

Pricing

Input price
$0.00/ 1M tokens
Output price
$0.00/ 1M tokens
Context window 262.1K tokensCompatible endpoints openaiVendor Alibaba

Performance

Loading performance data...

Usage & Ranking

Loading usage...

Supported parameters

All providers = every upstream serving this model supports it. Some providers = depends on which upstream handles the request. Default = the value sent when you leave the parameter unset.

ParameterProvidersDefault
frequency_penaltySome providersNot sent by default
include_reasoningAll providers-
logit_biasSome providers-
logprobsSome providers-
max_tokensAll providers-
min_pSome providers-
presence_penaltyAll providers-
reasoningAll providers-
repetition_penaltySome providers-
response_formatSome providers-
seedAll providers-
stopSome providers-
temperatureAll providersNot sent by default
tool_choiceSome providers-
toolsSome providers-
top_kSome providers-
top_logprobsSome providers-
top_pAll providersNot sent by default

Frequently asked questions

Similar models