docs: improve local backend reliability and documentation

- Switch local endpoints from localhost to 127.0.0.1 to avoid IPv6 resolution issues\n- Add recommended Ollama user systemd service example with keep-alive and CUDA support

README.md

@@ -448,7 +448,7 @@ audio:
   enabled: true
   provider:
     type: custom                            # openai, groq, ollama, llamacpp, custom
-    endpoint: "http://localhost:18801/v1/audio/transcriptions"
+    endpoint: "http://127.0.0.1:18801/v1/audio/transcriptions"
     api_key: "${WHISPER_API_KEY}"            # Optional Bearer token
     model: "whisper-1"                       # Model name (default: whisper-1)
 ```
@@ -549,7 +549,7 @@ docker compose --profile search up -d searxng
 Local-first + homelab fallback pattern:
 
 - If Flynn runs in Docker Compose, set `endpoint: "http://searxng:8080"`.
-- If Flynn runs on the host (`pnpm dev` / `pnpm start`), set `endpoint: "http://localhost:18803"`.
+- If Flynn runs on the host (`pnpm dev` / `pnpm start`), set `endpoint: "http://127.0.0.1:18803"`.
 - Set `fallback_endpoint` to your existing homelab SearXNG URL.
 
 Example for host-run Flynn with local compose SearXNG and homelab backup:
@@ -557,7 +557,7 @@ Example for host-run Flynn with local compose SearXNG and homelab backup:
 ```yaml
 web_search:
   provider: searxng
-  endpoint: "http://localhost:18803"
+  endpoint: "http://127.0.0.1:18803"
   fallback_endpoint: "https://searxng.homelab.local"
   max_results: 5
 ```
@@ -795,6 +795,37 @@ Or via tools:
 
 ## Running as Service
 
+### Ollama (Recommended)
+
+If running Ollama locally, use a user-level service to ensure it stays alive and reachable:
+
+```bash
+cat > ~/.config/systemd/user/ollama.service << 'EOF'
+[Unit]
+Description=Ollama Service (User)
+After=network.target
+
+[Service]
+Type=simple
+ExecStart=/usr/bin/ollama serve
+Restart=always
+RestartSec=3
+Environment="HOME=%h"
+Environment="OLLAMA_MODELS=%h/.ollama"
+Environment="LD_LIBRARY_PATH=/usr/lib:/opt/cuda/lib64"
+Environment="CUDA_VISIBLE_DEVICES=0"
+Environment="OLLAMA_KEEP_ALIVE=-1"
+
+[Install]
+WantedBy=default.target
+EOF
+
+systemctl --user daemon-reload
+systemctl --user enable --now ollama
+```
+
+### Flynn daemon
+
 ```bash
 # Create systemd user service
 mkdir -p ~/.config/systemd/user
@@ -1020,7 +1051,7 @@ backup:
   include_vectors: true
   minio:
     enabled: true
-    endpoint: localhost:9000
+    endpoint: 127.0.0.1:9000
     access_key: "${MINIO_ACCESS_KEY}"
     secret_key: "${MINIO_SECRET_KEY}"
     bucket: flynn-backups
@@ -1411,7 +1442,7 @@ memory:
 |----------|--------|
 | OpenAI | `provider: openai`, `api_key`, `model` (default: `text-embedding-3-small`) |
 | Gemini | `provider: gemini`, `api_key`, `model` |
-| Ollama | `provider: ollama`, `endpoint` (default: localhost:11434), `model` |
+| Ollama | `provider: ollama`, `endpoint` (default: 127.0.0.1:11434), `model` |
 | llama.cpp | `provider: llamacpp`, `endpoint`, optional `model` |
 | Voyage AI | `provider: voyageai`, `api_key` or `VOYAGE_API_KEY`, `model` (default: `voyage-3-large`) |
 

config/default.yaml

@@ -186,11 +186,11 @@ models:
     ollama:
       provider: ollama
       model: glm-4.7-flash
-      endpoint: http://localhost:11434
+      endpoint: http://127.0.0.1:11434
     llamacpp:
       provider: llamacpp
       model: gpt-oss-20b
-      endpoint: http://localhost:8080
+      endpoint: http://127.0.0.1:8080
   #
   # Then reference them in fallback_chain:
   #   fallback_chain: [ollama, llamacpp, local]
@@ -311,7 +311,7 @@ memory:
     enabled: true
     provider: ollama
     model: nomic-embed-text
-    endpoint: http://localhost:11434
+    endpoint: http://127.0.0.1:11434
     chunk_size: 512
     chunk_overlap: 50
     top_k: 5
@@ -565,7 +565,7 @@ automation:
 #   enabled: true
 #   provider:
 #     type: custom                            # openai, groq, ollama, llamacpp, custom
-#     endpoint: "http://localhost:18801/v1/audio/transcriptions"
+#     endpoint: "http://127.0.0.1:18801/v1/audio/transcriptions"
 #     api_key: "${WHISPER_API_KEY}"            # Optional Bearer token
 #     model: "whisper-1"                       # Model name (default: whisper-1)
 #   talk_mode: