---
id: "qwen/qwen3-235b-a22b-thinking-2507"
also: ["qwen/qwen3-235b-a22b-thinking-2507", "Qwen/Qwen3-235B-A22B-Thinking-2507", "hf/Qwen/Qwen3-235B-A22B-Thinking-2507", "requesty/alibaba/qwen3-235b-a22b-thinking-2507"]
cheapest_list_usd_per_m: "0.23 in, 2.3 out, at deepinfra on hf"
cheapest_effective_usd_per_m: "0.2248 in, 2.3 out, at deepinfra on requesty"
context: 128000
max_output: 16384
input: ["text"]
output: ["text"]
released: "2025-07-25"
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logit_bias", "logprobs", "max_tokens", "min_p", "presence_penalty", "reasoning", "repetition_penalty", "response_format", "seed", "stop", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 8
gateways: "openrouter (4), hf (3), requesty (1)"
no_training: 5
trains: 0
retains_prompts: 1
coding: 22.1000
agentic: 1.3000
---
# Qwen: Qwen3 235B A22B Thinking 2507

Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144...