Qwen3 235B

by Qwen · chat

Qwen3's flagship MoE model with 235B total / 22B active parameters. Supports dual thinking/non-thinking modes for complex reasoning, code generation, and multilingual tasks with 131K context.

Pricing

Input: $0.537143/M tokens · Output: $2.148571/M tokens

Capabilities

Function Calling, JSON Mode, Streaming, Reasoning

Context: 131K tokens

Max output: 8K tokens

Routes: 2/2 healthy

Performance

TTFT: 2797ms · Latency: 18645ms · Throughput: 30.8 tok/s