Qw
qwen3-30b-a3b-fp8 Benchmark & Insights
Qwen Cloudflare Workers AI
Updated Feb 20, 2026 All models
Sample size
6 runs
in window
Accuracy
50.0%
consensus match · 6d
Confidence
93%
over 6 runs
Window end
Feb 20, 2026
most recent run
Input price
$0.05/MTok
prompt tokens
Output price
$0.34/MTok
completion tokens
Model insights
- 01 Said "medium" on every one of its 6 days regardless of weather — a coin-flip score that suggests it isn't reacting to the input at all.
- 02 Undersampled, but the stuck-output pattern is disqualifying unless it changes with more data.