LOT 1 - Optimizare script build -Instalare mono comanda

ai_platform/modules/embeddings/INDEX.md

@@ -20,7 +20,7 @@ Generate text embeddings for downstream DIDI consumers:
 
 - **didi-brain semantic atom storage** — text claims/atoms get embedded once at ingestion, stored as pgvector columns.
 - **didi-brain `/v1/gather` query embedding** — incoming claim text gets embedded, used as the kNN probe vector.
-- **Cross-encoder reranking** — separate `rerank` module (port 54200) handles re-scoring; this module only does dense embeddings.
+- **Cross-encoder reranking** — separate `rerank` module (port 14200) handles re-scoring; this module only does dense embeddings.
 - **Catalog / retrieval consumers** — anything that needs a vector representation hits this single endpoint.
 
 The module is OpenAI-compatible: drop-in for `openai` SDK code that uses `client.embeddings.create()`.
@@ -68,7 +68,6 @@ embeddings/
 ├── README.md                  quick-start + config table
 ├── API.md                     full HTTP API reference + SDK examples
 ├── pyproject.toml             package def, optional extras: vllm, llamacpp, all, dev
-├── uv.lock
 ├── .env.example               all EMB_* env vars documented
 ├── deploy/
 │   ├── Dockerfile             multi-stage, python 3.11-slim + uv
@@ -111,7 +110,7 @@ embeddings/
 
 - **Ingestion path** — when atoms/claims are written, brain calls `EmbeddingClient.embed(text)` to get a 1024-dim vector and stores it in pgvector alongside the row. One round-trip per batch.
 - **Query path** — for `/v1/gather`, the incoming claim text is embedded the same way, then the resulting vector is used as the probe in a pgvector `<=> ` (cosine) kNN search to retrieve candidate atoms.
-- **Reranking** — top-k candidates from the dense search are forwarded to the separate `rerank` module (cross-encoder, port 54200) for fine-grained scoring. That module does not call this one — they're parallel concerns.
+- **Reranking** — top-k candidates from the dense search are forwarded to the separate `rerank` module (cross-encoder, port 14200) for fine-grained scoring. That module does not call this one — they're parallel concerns.
 - **Catalog API** — also retrieves via the same embedding pipeline (text → vector → kNN), reusing this single endpoint.
 
 Because the wrapper is OpenAI-compatible, `EmbeddingClient` can be a vanilla `openai.OpenAI(base_url=..., api_key=...)` instance — no DIDI-specific client code needed in brain.
@@ -138,16 +137,9 @@ All env vars use the `EMB_` prefix. Required vars have no defaults — the app r
 | `EMB_PORT` | `14100` (Prod) / `54100` (Dev) | API server port |
 | `EMB_HOST` | `0.0.0.0` | API bind address |
 | `EMB_API_TOKENS` | unset | Comma-separated bearer tokens; auth disabled if unset |
-| `EMB_VLLM_BASE_URL` | `http://localhost:54101` | Where vLLM listens |
-| `EMB_VLLM_MODEL` | `BAAI/bge-m3` | HF model id loaded by vLLM |
-| `EMB_VLLM_GPU` | `0` | `CUDA_VISIBLE_DEVICES` for vLLM |
-| `EMB_VLLM_GPU_UTIL` | `0.50` | vLLM `--gpu-memory-utilization` |
-| `EMB_VLLM_MAX_LEN` | `8192` | vLLM `--max-model-len` |
-| `EMB_LLAMACPP_BASE_URL` | `http://localhost:54110` | Where llama.cpp listens |
-| `EMB_LLAMACPP_MODEL` | `bge-m3-q4_k_m.gguf` | GGUF filename inside `MODELS_DIR` |
-| `EMB_LLAMACPP_CTX` | `8192` | llama.cpp context size |
-| `EMB_LLAMACPP_THREADS` | `4` | llama.cpp threads |
-| `EMB_LLAMACPP_PARALLEL` | `4` | llama.cpp parallel slots |
+| `EMB_VLLM_BASE_URL` | `http://localhost:54101` | Where the API wrapper reaches vLLM |
+| `EMB_VLLM_API_KEY` | unset | Optional API key for the vLLM server |
+| `EMB_LLAMACPP_BASE_URL` | `http://localhost:54110` | Where the API wrapper reaches llama.cpp |
 | `EMB_REQUEST_TIMEOUT` | `120.0` | Per-request backend timeout (s) |
 | `EMB_CONNECT_TIMEOUT` | `10.0` | TCP connect timeout (s) |
 | `EMB_RATE_LIMIT_RPS` | `20.0` | Token-bucket rate (req/s) |
@@ -155,7 +147,24 @@ All env vars use the `EMB_` prefix. Required vars have no defaults — the app r
 | `EMB_MAX_CONCURRENT_REQUESTS` | `20` | In-flight cap |
 | `EMB_LOG_LEVEL` | `INFO` | DEBUG / INFO / WARNING / ERROR |
 | `EMB_LOG_JSON` | `false` | Emit JSON-formatted log lines |
-| `HF_CACHE_DIR` | `/cai2_ds_storage/hf_cache` | Mounted into vLLM container |
+
+> The API app reads only the variables above (fields on `EmbeddingSettings` in `config.py`). `EMB_VLLM_BASE_URL` / `EMB_LLAMACPP_BASE_URL` just tell the wrapper where to reach the backend servers.
+
+### Backend service variables (docker-compose only)
+
+These are **not** fields on `EmbeddingSettings` — they are consumed by `deploy/docker-compose.yml` to launch the vLLM / llama.cpp containers (model id, GPU pinning, memory, context). They configure the backend server, not the API app.
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `EMB_VLLM_MODEL` | `BAAI/bge-m3` | HF model id loaded by the vLLM container |
+| `EMB_VLLM_GPU` | `0` | `CUDA_VISIBLE_DEVICES` for the vLLM container |
+| `EMB_VLLM_GPU_UTIL` | `0.50` | vLLM `--gpu-memory-utilization` |
+| `EMB_VLLM_MAX_LEN` | `8192` | vLLM `--max-model-len` |
+| `EMB_LLAMACPP_MODEL` | `bge-m3-q4_k_m.gguf` | GGUF filename inside `MODELS_DIR` |
+| `EMB_LLAMACPP_CTX` | `8192` | llama.cpp context size |
+| `EMB_LLAMACPP_THREADS` | `4` | llama.cpp threads |
+| `EMB_LLAMACPP_PARALLEL` | `4` | llama.cpp parallel slots |
+| `HF_CACHE_DIR` | `/cai2_ds_storage/hf_cache` | Mounted into the vLLM container |
 | `HF_TOKEN` | unset | Forwarded as `HUGGING_FACE_HUB_TOKEN` |
 | `MODELS_DIR` | `/cai2_ds_storage/models` | GGUF model directory for llama.cpp |
 
@@ -208,7 +217,7 @@ Compose profiles:
 ## Related
 
 - **`didi-brain`** consumes via `shared/embedding_client.py` for both ingestion (text → pgvector storage) and `/v1/gather` query embedding.
-- **`rerank` module** (separate, port 54200) handles cross-encoder scoring on top of dense kNN candidates from this module — they're complementary, not chained inside this service.
+- **`rerank` module** (separate, port 14200) handles cross-encoder scoring on top of dense kNN candidates from this module — they're complementary, not chained inside this service.
 - **Catalog API** uses the same endpoint for retrieval-side embeddings.
 - **pgvector** in the brain Postgres stores the resulting 1024-dim vectors (cosine distance index).
 - **Local Python use**: `from embeddings import EmbeddingClient` (works without the HTTP wrapper if you want in-process inference and have the `vllm`/`llamacpp` extras installed).

ai_platform/modules/embeddings/deploy/docker-compose.yml

@@ -14,10 +14,10 @@
 # Naming Convention: didiAI-{module}-{service}
 #
 # Network:
-#   Uses deploy_default network (shared with other modules)
+#   Uses didi-network (shared with all DIDI + AI platform stacks)
 
 networks:
-  deploy_default:
+  didi-network:
     external: true
 
 services:
@@ -33,7 +33,7 @@ services:
     ports:
       - "${EMB_PORT:-14100}:${EMB_PORT:-14100}"
     networks:
-      - deploy_default
+      - didi-network
     environment:
       - EMB_PORT=${EMB_PORT:-14100}
       - EMB_EXTERNAL_URL=${EMB_EXTERNAL_URL}
@@ -66,7 +66,7 @@ services:
     ports:
       - "${EMB_VLLM_PORT:-14101}:14101"
     networks:
-      - deploy_default
+      - didi-network
     volumes:
       - ${HF_CACHE_DIR:-/cai2_ds_storage/hf_cache}:/root/.cache/huggingface
     environment:
@@ -109,7 +109,7 @@ services:
     ports:
       - "${EMB_LLAMACPP_PORT:-14110}:8080"
     networks:
-      - deploy_default
+      - didi-network
     volumes:
       - ${MODELS_DIR:-/cai2_ds_storage/models}:/models:ro
     command: >