Qwen3.8 27B

qwen-3.8-27b

Dense multimodal model for agentic coding and tool use, served at wafer-scale speed

Intelligence: 33.7
Context: 131K
Training cutoff: --
Added: 2026-09-04
Inputs: Vision

Providers

One route serves Qwen3.8 27B, and every figure below is measured on it. Where the selected window has too little traffic, the figure names what it is based on instead.

Cerebras only route

qwen-3.8-27b

Uptime

98.9%

Input / 1M: $0.99
Output / 1M: $1.49
TTFT: 0.93s live traffic · 7d · n=357
Output rate: 265.6 tok/s live traffic · 7d · n=357

131K context 40,960 max output cache $0.99 read / $0.99 write per 1M Vision ZDRwarming · 30m

TTFT is the median time to the first visible answer token over real ModelRelay requests, with a fixed prompt standing in for models that have no traffic yet. Speed ranks by it. Intelligence is the Artificial Analysis Intelligence Index.