fix: keep local embedding batches from flooding providers

docs/.generated/config-baseline.sha256

@@ -1,4 +1,4 @@
-0b0d796bceddfb9e2929518ba84af626da7f5d75c392a217041f36e850c4e74f  config-baseline.json
-271fdf1d6652927e0fc160a6f25276bf6dccb8f1b27fab15e0fc2620e8cacab4  config-baseline.core.json
+3b9a8841973205560a5396e7a18d301852941a95a561900984ad618e69a99d05  config-baseline.json
+089ab9493c8482687f19da89d37e069fc402543696c92e6e3be86072c1e48c68  config-baseline.core.json
 7cd9c908f066c143eab2a201efbc9640f483ab28bba92ddeca1d18cc2b528bc3  config-baseline.channel.json
 17eb3f8887193579ff32e35f9bd520ba2bd6049e52ab18855c5d41fcbf195d83  config-baseline.plugin.json

docs/concepts/memory-search.md

@@ -135,6 +135,11 @@ earlier conversations. This is opt-in via
 **Only keyword matches?** Your embedding provider may not be configured. Check
 `openclaw memory status --deep`.
 
+**Local embeddings time out?** `ollama`, `lmstudio`, and `local` use a longer
+inline batch timeout by default. If the host is simply slow, set
+`agents.defaults.memorySearch.sync.embeddingBatchTimeoutSeconds` and rerun
+`openclaw memory index --force`.
+
 **CJK text not found?** Rebuild the FTS index with
 `openclaw memory index --force`.
 

docs/reference/memory-config.md

@@ -219,6 +219,17 @@ to an existing local file. `hf:` and HTTP(S) model references can still be used
 explicitly with `provider: "local"`, but they do not make `auto` select local
 before the model is available on disk.
 
+### Inline embedding timeout
+
+| Key                                 | Type     | Default          | Description                                                              |
+| ----------------------------------- | -------- | ---------------- | ------------------------------------------------------------------------ |
+| `sync.embeddingBatchTimeoutSeconds` | `number` | provider default | Override the timeout for inline embedding batches during memory indexing |
+
+Unset uses the provider default: 600 seconds for local/self-hosted providers
+such as `local`, `ollama`, and `lmstudio`, and 120 seconds for hosted providers.
+
+Increase this when local CPU-bound embedding batches are healthy but slow.
+
 ---
 
 ## Hybrid search config
@@ -347,6 +358,10 @@ Prevents re-embedding unchanged text during reindex or transcript updates.
 Available for `openai`, `gemini`, and `voyage`. OpenAI batch is typically
 fastest and cheapest for large backfills.
 
+This is separate from `sync.embeddingBatchTimeoutSeconds`, which controls inline
+embedding calls used by local/self-hosted providers and hosted providers when
+provider batch APIs are not active.
+
 ---
 
 ## Session memory search (experimental)