Documentation
¶
Overview ¶
Package embedding provides text embedding capabilities.
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func ProbeV2 ¶ added in v0.35.0
ProbeV2 reports whether the proxy exposes the versioned /v2/embedding route and, if so, the embedding model it currently advertises. It POSTs an empty check request to /v2/embedding/check; a 200 response means v2 is available and its model is returned. Any non-200 status or transport error yields ("", false), signalling the caller to fall back to the legacy /embed routes.
Types ¶
type Embedder ¶
type Embedder interface {
// Embed returns the L2-normalized embedding vector for a single text string.
Embed(text string) ([]float32, error)
// EmbedBatch returns L2-normalized embedding vectors for multiple texts.
EmbedBatch(texts []string) ([][]float32, error)
// Close releases resources held by the embedder.
Close() error
}
Embedder provides text embedding capabilities.
type RemoteEmbedder ¶ added in v0.14.0
type RemoteEmbedder struct {
// contains filtered or unexported fields
}
RemoteEmbedder implements Embedder by calling the proxy's embed endpoint. An optional local cache avoids round-trips to the proxy on warm restarts. When v2 is set it targets the versioned /v2/embedding routes (fp32 at a fixed dimensionality, model advertised per response); otherwise it uses the legacy /embed routes.
func NewRemote ¶ added in v0.14.0
func NewRemote( log logrus.FieldLogger, proxyURL string, tokenFn func() string, invalidateFn func(), localCache cache.Cache, model string, ) *RemoteEmbedder
NewRemote creates a RemoteEmbedder that calls the proxy's legacy /embed routes. tokenFn is called on each request to get the current auth token, and invalidateFn drops the cached token so a 401/403 can be retried with a fresh one (it may be nil to disable the retry). localCache and model are optional — when both are set, embedding vectors are cached locally using {model}:{textHash} keys to avoid proxy round-trips.
func NewRemoteWithEndpoint ¶ added in v0.35.0
func NewRemoteWithEndpoint( log logrus.FieldLogger, proxyURL string, tokenFn func() string, invalidateFn func(), localCache cache.Cache, model string, v2 bool, ) *RemoteEmbedder
NewRemoteWithEndpoint creates a RemoteEmbedder targeting either the v2 (/v2/embedding) routes when v2 is true, or the legacy /embed routes otherwise.
func (*RemoteEmbedder) Close ¶ added in v0.14.0
func (e *RemoteEmbedder) Close() error
Close releases resources held by the embedder.
func (*RemoteEmbedder) Embed ¶ added in v0.14.0
func (e *RemoteEmbedder) Embed(text string) ([]float32, error)
Embed returns the L2-normalized embedding vector for a single text string.
func (*RemoteEmbedder) EmbedBatch ¶ added in v0.14.0
func (e *RemoteEmbedder) EmbedBatch(texts []string) ([][]float32, error)
EmbedBatch returns L2-normalized embedding vectors for multiple texts. When a local cache is configured, vectors are checked there first. Remaining misses are split into sub-batches of maxBatchSize and sent to the proxy (which has its own Redis cache + upstream API).
func (*RemoteEmbedder) Model ¶ added in v0.35.0
func (e *RemoteEmbedder) Model() string
Model returns the embedding model this embedder is keyed to. Index builds use it to tag the embedding space they were built in, so a later change to the proxy's served model can be detected and trigger a re-index.
func (*RemoteEmbedder) OnProgress ¶ added in v0.34.0
func (e *RemoteEmbedder) OnProgress(fn func(completed, total int))
OnProgress registers a callback invoked during EmbedBatch with the number of documents embedded so far and the total in the batch. It enables document-level progress reporting for index builds.
Source Files