Apps & Tools
Opus-5.5 Inference Optimization
Performance-optimized Docker image for Opus-5.5 reaching 500+ tok/s decode
by0xSero
Is this your creation?

Description
A high-performance inference optimization for Claude Opus 5.5 that achieves a 39% improvement in single-user decode speed on 4x H200 hardware. The build significantly boosts prefill throughput to 18-33k tok/s, representing a 1.6x-1.9x gain over previous versions.
Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.