News for hackers. Stuff that matters.

A comment on DeepSeek: X2 Speed for WASM with SIMD. Parent comment

Re:DeepSeek: X2 Speed for WASM with SIMD by lenerdenator on Tuesday January 28, 2025 @11:19AM (#42854087)
My greatest problem is duplicating the secret sauce of GHCP: it has access to your project and can use it as context.

Admittedly, I haven't looked too hard, but how could I do that with a model from, say, Ollama and run exclusively on my machine?

[ Parent | on HN ]
Re:DeepSeek: X2 Speed for WASM with SIMD by simonw on Tuesday January 28, 2025 @11:21AM (#42854113)
There are a bunch of tools that might be able to do that. I'd start by exploring https://aider.chat/
[ Parent | on HN ]
Re:DeepSeek: X2 Speed for WASM with SIMD by realo on Tuesday January 28, 2025 @12:00PM (#42854702)
Neat. Can I use aider with a local model running in LMStudio (or ollama)?

After a very quick reading of their pages it does not seem so.

Hopefully I am wrong...

[ Parent | on HN ]
Re:DeepSeek: X2 Speed for WASM with SIMD by simonw on Tuesday January 28, 2025 @12:01PM (#42854708)
Yes you can: https://aider.chat/docs/llms/ollama.html
[ Parent | on HN ]
Re:DeepSeek: X2 Speed for WASM with SIMD by lenerdenator on Tuesday January 28, 2025 @11:43AM (#42854438)
I'll take a look, thank you!
[ Parent | on HN ]
Re:DeepSeek: X2 Speed for WASM with SIMD by nkozyra on Tuesday January 28, 2025 @11:51AM (#42854563)
Couldn't you load the whole thing into a database or memory and use it as a RAG source? Not sure if that would fully scratch the itch.
[ Parent | on HN ]