Qwen 3.6 35B A3B

by Qwen · chat

Qwen 3.6 开源 MoE 模型(35B 总参 / 3B 激活,Apache 2.0)。262K 原生上下文。支持视觉+工具+推理。

Pricing

Input: $0.249635/M tokens · Output: .49781/M tokens

Capabilities

Vision, Function Calling, JSON Mode, Streaming, Reasoning

Context: 262K tokens

Max output: 82K tokens

Routes: 1/1 healthy

Performance

TTFT: 528ms · Latency: 3519ms · Throughput: 53.1 tok/s