AI-assissted firmware analysisEMBAenhancementgood first issuehelp wanted
Métriques du dépôt
- Stars
- (3 565 étoiles)
- Métriques de merge PR
- (Métriques PR en attente)
Description
Is your feature request related to a problem? Please describe.
Please expand the options in your LocalAI calls to allow us to configure more elements of the URL.
Currently, the Scheme, Port and API_endpoint are hardcoded in the /modules/Q03_localai_connector.sh file:
- line 292: http://"${LOCAL_AI_IP}":8080/v1/chat/completions
- line 419: http://"${LOCAL_AI_IP}":8080/v1/models
Describe the solution you'd like
- It would be nice if the
AI_OPTION=3parameter could be set as a command line option and/or fixed in the/config/ai_config.envfile. - In the
/config/ai_config.envfile contains the
- Schema (HTTP/HTTPS) -> LOCAL_AI_SCHEME="HTTPS"
- Port (0-65535) -> LOCAL_AI_PORT="8080"
- Path to API endpoint: -> LOCAL_AI_API_ENDPOINT="/v1"
- https://HOSTNAME/v1 - works with the OpenAI libraries and any OpenAI-compatible tool
- https://HOSTNAME/llm - works with Claude Code and other Anthropic tools
Describe alternatives you've considered Instead, we could use an HTTP proxy.
Priority issue Are you already a Sponsor? - [Y/N] N
Additional context
