Local AI Cat module
MLXServe embedded listener
Current shipping, testing, and product context for this part of Local AI Cat.
Status
Where it stands
- Platforms
- iOS · macOS
- Release channels
- Alpha · Beta · Main
- Distribution
- Indoor · Outdoor
- AlphaiOS (partial) · macOS (partial)
- BetaiOS (partial) · macOS (partial)
- Outdoor · Direct downloadiOS (partial) · macOS (partial)
- Indoor · App StoreiOS (partial) · macOS (partial)
Description
What it does
Cross-platform default-off A/B infrastructure. Adds FEATURE_MLXSERVE_EMBED and links only the MLXServeHTTP product from local-ai-cat/mlxserve. The adapter implements mlxserve's OpenAIChatBackend over the app's existing GenerationQueue/TextGenerationRuntime lineage, preserving the single resident model runtime invariant. Start and termination are registered through LaunchParticipant and receive the app runtime's shared GenerationQueue handle through injection. macOS keeps the out-of-process sidecar as the default A/B path; iOS can opt into an in-process loopback listener on port 11435 for device validation.
Headless Local API
Drive it over the loopback API
Logging / observability: provisional grade L0-L1 — app-target — see logging audit.
- discover
- not applicable
- execute
- present
- observe
- not applicable
- configure
- not applicable
- cancel/stop
- not applicable
- reset
- not applicable
/v1/local/app/action/v1/local/app/artifacts/{id}/v1/local/app/navigate/v1/local/app/screenshot/v1/local/app/state/v1/local/app/surfaces/v1/local/app/transactions/{id}/v1/local/artifacts/v1/local/artifacts/{artifact_id}/v1/local/artifacts/{artifact_id}/content
Media
See it in action
Backlog / improvements
What comes next
No improvements are listed for this module yet.
History
How it got here
No change history has been published for this module yet.