You can also try speculative decoding with the E2B model. Under some conditions ... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

Havoc 5 months ago | parent | context | favorite | on: I ran Gemma 4 as a local model in Codex CLI

You can also try speculative decoding with the E2B model. Under some conditions it can result in a decent speed up

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact