Pre-existing issue: cmd/cmd_benchmark.go imports go-i18n/reversal
but the dependency was absent from go.mod.
Co-Authored-By: Virgil <virgil@lethean.io>
Move all 40 ML command files from core/cli/cmd/ml to go-ml/cmd/.
Commands self-register via init() + cli.RegisterCommands().
- Fixed stale go-ai/ml import in cmd_ab.go (now uses go-ml + go-mlx)
- Disabled cmd_train.go (needs go-mlx training API export)
- Removed all local replace directives, deps resolve from forge
- go build ./... passes cleanly
Co-Authored-By: Virgil <virgil@lethean.io>
Implements RouteGroup and StreamGroup interfaces from go-api, exposing
ML inference service methods as REST endpoints: GET /backends, GET /status,
POST /generate. Includes 12 integration tests covering interface satisfaction,
engine registration, envelope format, and error handling.
Co-Authored-By: Virgil <virgil@lethean.io>
InferenceAdapter wraps inference.TextModel (iter.Seq[Token]) to satisfy
ml.Backend (string returns) and ml.StreamingBackend (TokenCallback).
- adapter.go: InferenceAdapter with Generate/Chat/Stream/Close
- adapter_test.go: 13 test cases with mock TextModel (all pass)
- backend_mlx.go: rewritten from 253 LOC to ~35 LOC using go-inference
- go.mod: add forge.lthn.ai/core/go-inference dependency
- TODO.md: mark Phase 1 steps 1.1-1.3 complete
Co-Authored-By: Virgil <virgil@lethean.io>
