Switch to bge-base-en-v1.5 (512 token context)

README.md

@@ -6,7 +6,7 @@ Lightweight embedding service for Clawdbot memory search.
 
 Runs [HuggingFace Text Embeddings Inference](https://github.com/huggingface/text-embeddings-inference) (TEI) on CPU to provide OpenAI-compatible embeddings for semantic memory search.
 
-- **Model:** `sentence-transformers/all-MiniLM-L6-v2` (~90MB, 384 dimensions)
+- **Model:** `BAAI/bge-base-en-v1.5` (~440MB, 768 dimensions, 512 token context)
 - **Endpoint:** `http://text-embeddings:8080/v1/embeddings`
 - **No GPU required**
 
@@ -28,7 +28,7 @@ Add to `~/.clawdbot/clawdbot.json`:
       "memorySearch": {
         "enabled": true,
         "provider": "openai",
-        "model": "sentence-transformers/all-MiniLM-L6-v2",
+        "model": "BAAI/bge-base-en-v1.5",
         "remote": {
           "baseUrl": "http://text-embeddings.tei.svc.cluster.local:8080/v1/",
           "apiKey": "not-needed"
@@ -50,5 +50,5 @@ Add to `~/.clawdbot/clawdbot.json`:
 ```bash
 curl -X POST http://text-embeddings:8080/v1/embeddings \
   -H "Content-Type: application/json" \
-  -d '{"input": "Hello world", "model": "sentence-transformers/all-MiniLM-L6-v2"}'
+  -d '{"input": "Hello world", "model": "BAAI/bge-base-en-v1.5"}'
 ```

deployment.yaml

@@ -33,7 +33,7 @@ spec:
           image: ghcr.io/huggingface/text-embeddings-inference:cpu-1.6
           args:
             - --model-id
-            - sentence-transformers/all-MiniLM-L6-v2
+            - BAAI/bge-base-en-v1.5
             - --port
             - "8080"
           ports: