Conversation
Add gemini-embedding-2-preview to the known model registry with multimodal capabilities (images, audio, video, PDFs, 3072 dims, fusion support). Introduce `multimodal: true` config escape hatch so users can declare any future model as multimodal without waiting for a registry update. The GenaiGoogleImpl (gemini provider) now routes multimodal content through the genai SDK's NewPartFromBytes/NewContentFromParts path. Also: - Wire GetConfigCapabilities() through all providers (OpenAI, Bedrock, Ollama, Cohere, OpenRouter, Termite) - Use capabilities-based routing in vertex provider instead of hardcoded model name check - Guard vertex provider's multimodal path to only allow multimodalembedding models (legacy Prediction API), with clear error pointing to gemini provider for other models - Fix copy-paste "ollama" error message in GenaiGoogleImpl.GetModels() - Simplify: merge config_capabilities.go into capabilities.go, reuse allText() in vertex.go, clean up NewTermiteClient signature - Add integration tests (skip without credentials) and capability tests
Picks up fix for nil map panic in zapx invertedIndexCache.Clear() and synonymIndexCache.Clear() during concurrent shard split operations.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
gemini-embedding-2-previewto the known model registry with multimodal capabilities (images, audio, video, PDFs, 3072 dims, fusion support)multimodal: trueconfig escape hatch so users can declare any future model as multimodal without waiting for a registry updategoogle_gemini.go) to route multimodal content through the genai SDK'sNewPartFromBytes/NewContentFromPartspathGetConfigCapabilities()through all providers (OpenAI, Bedrock, Ollama, Cohere, OpenRouter, Termite)multimodalembeddingmodels, with a clear error directing users to useprovider: "gemini"withGOOGLE_GENAI_USE_VERTEXAI=1for other modelsGenaiGoogleImpl.GetModels()config_capabilities.gointocapabilities.go, reuseallText()invertex.go, clean upNewTermiteClientsignatureTest plan
lib/embeddings/...tests pass locallyGEMINI_API_KEYset (text, image, fused text+image)