Models / Qwen

Qwen 3.8 27B

qwen3.8-27b

Fast 27B-class multimodal model with automatic prompt caching. Cached input never costs more than fresh input.

Lite · Pro · Max · Team · Kenya Weekly 1,001,536 token context

Plans and limits

Estimated requests per 5 hours, assuming 6,000 input and 800 output tokens per request. The enforced limit is each plan's dollar allowance.

Lite

~909

requests per 5 hours

Pro

~2,727

requests per 5 hours

Max

~9,090

requests per 5 hours

Pay as you go: $0.045 input and $0.350 output per million tokens.

Use it in Fikra Code

$ fikra -m qwen3.8-27b

Or type /model inside a session and pick it from the list.

See plans