DeepSeek's multi-head latent attention and other KV cache tricks

Article URL: https://garymarcus.substack.com/p/hot-take-gpt-45-is-a-nothing-burger

Comments URL:

Article URL: https://www.elenaverna.com/p/ai-is-killing-some-companies-yet

Comments URL:

Article URL: https://azsh.it

Comments URL: https://news.ycombinator.com/item?id=43210536

Poi

Article URL: https://www.construction-physics.com/p/why-its-so-hard-to-build-a-jet-engine

Comments

Article URL: https://github.com/cmackenzie1/torii-rs

Comments URL: https://news

Article URL: https://www.economist.com/leaders/2025/02/27/inheriting-is-becoming-nearly

Some background: I work on Langfuse and we've been collaborating with LiteLLM.

(LiteLLM is a Python library and proxy/gateway that handles cost management, virtual keys, caching, and rate-limiti