Matthew Farrellee

i setup with: `pass init && pass git init` my git url is: _ssh://me@home/home/me/**.password-store**_ i'm using an ssh key _(aside: ssh key had no password, but i was still prompted...

fyi - https://app.netlify.com/sites/stenographer-chimpanzee-84661/deploys/5bb4d85c1f12b75ba2e69e82

@jkremser does it work for you?

an alternative way to detect driver presence and version: `modinfo -F version nvidia`

> The different env var is necessary since we're testing different services that need different API keys. the same api key should work for integrate.api.nvidia.com, ai.api.nvidia.com and api.nvcf.nvidia.com

@dolfim-ibm i missed the fact you'd already started this. an alternative is in https://github.com/meta-llama/llama-stack/pull/2179

iirc, developing a RAG pipeline that used hybrid search. rank-bm25 was buried under a few frameworks, app -> framework+ -> rank-bm25, making the error difficult to interpret and patch.

that sounds like a great solution.

@jeffmaury which inference provider?

@jeffmaury where can i find the code? is the issue that users cannot register models not in the static list? is the issue that users should not need to register...