0xPlaygrounds / 0xPlaygrounds/rig
feat: Use gRPC APIs for Gemini + xAI
- Vorherrschende Sprache
- Rust
- Sterne
- 8.6k
- Forks
- 959
- Ø Merge
- 4 Std. 32 Min.
- Gemergte PRs (30 T.)
- 117
Beschreibung
## Feature Request
Use the gRPC API for the Rig model providers that expose gRPC APIs (currently Google (Gemini Direct API) and xAI)
Docs + proto sources:
- Google Gemini Direct API (REST/gRPC) docs:
- https://ai.google.dev/gemini-api/docs/partner-integration
- Google proto source (Gemini / Generative Language API):
- https://github.com/googleapis/googleapis/blob/master/google/ai/generativelanguage/v1beta/generative_service.proto
- xAI gRPC docs + proto sources:
- https://docs.x.ai/docs/grpc-reference
- https://github.com/xai-org/xai-proto
- https://github.com/xai-org/xai-proto/blob/736b835b0c0dd93698664732daad49f87a2fbc6f/proto/xai/api/v1/chat.proto
### Motivation
- Protobuf is strongly typed, more performant to de/serialize than JSON, smaller than JSON, easy to generate via the `tonic` crate and backwards compatible
### Proposal
- Use the gRPC backend for the existing `gemini` and `xai` providers instead of the current REST backend
- Implementation outline:
- Use `tonic` + `prost` (or equivalent) to generate clients from the upstream protos.
- Google:
- Generate a client for `google.ai.generativelanguage.v1beta.GenerativeService` from the googleapis proto.
- Attach `x-goog-api-key` (and optionally `x-goog-api-client` for identification) metadata headers.
- xAI:
- Generate clients from `xai-org/xai-proto` and connect to `api.x.ai` with `Authorization: Bearer ` metadata.
- Map Rig’s existing request/response types to/from the proto messages; preserve existing streaming APIs (`stream_prompt`, etc.) by adapting gRPC streaming responses into Rig’s stream types.
Beitragsleitfaden
Rechercherichtung
The issue requires generating gRPC clients from the provided proto files for Google Gemini and xAI using the `tonic` and `prost` crates. Start by examining the existing `gemini` and `xai` provider implementations in the codebase to understand the current REST API integration. Then, map Rig's request/response types to the generated proto messages, ensuring streaming support is maintained. Testing will involve verifying the new gRPC connections work with the respective authentication headers.
Vom Indexierungsmodell aus dem Issue-Text verfasst.
Bewertung
- Tech-Stack
- grpc, rust
- Bereich
- ai-infra-agents, api, backend
- Issue-Typ
- Feature
- Schwierigkeit
- 4/5
- Geschätzter Aufwand
- 3-5 Tage
- Aktivitätsstatus
- Veraltet
- Klarheit
- Klar beschrieben
- Anfängerfreundlichkeit
- 40/100