Commit graph

9 commits

Author SHA1 Message Date
Ashwin Bharambe
5dce17668c Move run-*.yaml to templates/ so they can be packaged 2024-11-18 14:54:20 -08:00
Ashwin Bharambe
1ecaf2cb3c Add ollama/pull-models.sh 2024-11-18 11:44:03 -08:00
Ashwin Bharambe
fa1d29cfdc kill built_at field in run config 2024-11-17 20:42:11 -08:00
Ashwin Bharambe
b1d119466e Allow setting environment variables from llama stack run and fix ollama 2024-11-17 19:49:15 -08:00
Ashwin Bharambe
a061f3f8c1 Convert ollama to the new model 2024-11-17 15:19:55 -08:00
Ashwin Bharambe
028530546f Convert TGI 2024-11-17 14:49:41 -08:00
Ashwin Bharambe
3d7561e55c
Rename all inline providers with an inline:: prefix (#423) 2024-11-11 22:19:16 -08:00
Ashwin Bharambe
c1f7ba3aed
Split safety into (llama-guard, prompt-guard, code-scanner) (#400)
Splits the meta-reference safety implementation into three distinct providers:

- inline::llama-guard
- inline::prompt-guard
- inline::code-scanner

Note that this PR is a backward incompatible change to the llama stack server. I have added deprecation_error field to ProviderSpec -- the server reads it and immediately barfs. This is used to direct the user with a specific message on what action to perform. An automagical "config upgrade" is a bit too much work to implement right now :/

(Note that we will be gradually prefixing all inline providers with inline:: -- I am only doing this for this set of new providers because otherwise existing configuration files will break even more badly.)
2024-11-11 09:29:18 -08:00
Xi Yan
b1d7376730 kill tgi/cpu 2024-11-08 10:33:45 -08:00
Renamed from distributions/tgi/gpu/run.yaml (Browse further)