HLS
Home
Traces
Blog
Inference Systems
/
Generic serving
Opus 5
max
on
Generic serving
Run 1
1.22×
Run 2
0
Run 3
0
Speedup
1.22×
Starting point
vLLM source
Agent
Claude Code
Build time
—
Steps
—
Tokens in / out
—
Cost
—
Loading the trajectory…