An AI-powered search engine with a generative UI
TypeScript generative-aigenerative-uinextjsollamareactsearxngshadcn-uitailwindcsstavilytypescriptupstashvercel-ai-sdk
docs: Simplify Ollama documentation and add disclaimer
Yoshiki Miurad64f8e83786f
diff --git a/config/models/cloud.json b/config/models/cloud.jsonindex e9ded07f..8bbc906a 100644--- a/config/models/cloud.json+++ b/config/models/cloud.json@@ -4,10 +4,10 @@"byMode": {"quick": {"speed": {- "id": "gpt-oss:20b",- "name": "GPT-OSS 20B",- "provider": "Ollama",- "providerId": "ollama"+ "id": "gemini-2.5-flash",+ "name": "Gemini 2.5 Flash",+ "provider": "Google",+ "providerId": "google"},"quality": {"id": "gpt-5-mini",diff --git a/docs/CONFIGURATION.md b/docs/CONFIGURATION.mdindex ec591c1e..624f412a 100644--- a/docs/CONFIGURATION.md+++ b/docs/CONFIGURATION.md@@ -195,34 +195,15 @@ AI_GATEWAY_API_KEY=[YOUR_AI_GATEWAY_API_KEY]#### Ollama (Self-hosted)-[Ollama](https://ollama.com/) enables you to run large language models locally on your own hardware, providing privacy, cost savings, and offline capabilities.+[Ollama](https://ollama.com/) enables you to run large language models locally on your own hardware.-**Setup:**--1. Install Ollama following the [official installation guide](https://ollama.com/download)--2. Pull a model with tools support (required for Morphic):--```bash-# Example: Pull Llama 3.2 with tools support-ollama pull llama3.2--# Or other models with tools capability:-ollama pull qwen2.5-ollama pull mistral-```--3. Configure your environment:+**Configuration:**```bashOLLAMA_BASE_URL=http://localhost:11434```-**Important Notes:**--- **Tools Capability Required**: Morphic requires models to support the `tools` capability for web search functionality. On server startup, Morphic will automatically check which of your Ollama models support tools and log the results.--- **Model Configuration**: Update your `config/models/*.json` files to use Ollama models. Set `providerId` to `ollama` and use the model name from Ollama:+Then update your `config/models/*.json` files to use Ollama models:```json{@@ -233,20 +214,15 @@ OLLAMA_BASE_URL=http://localhost:11434}```-- **Checking Model Capabilities**: To verify if a model supports tools, check the server logs on startup. Morphic will display which models passed validation:--```-✓ llama3.2:latest supports tools-✗ codellama:latest does not support tools (skipped)-```--- **Performance**: For best performance, ensure your system meets Ollama's hardware requirements. GPU acceleration is recommended for larger models.+**Important Notes:**-**Troubleshooting:**+- **Tools Capability**: Morphic requires models to support the `tools` capability for function calling. On server startup, Morphic validates configured models and logs the results. Note that even if a model reports tools support, actual tool calling performance depends on the model's capabilities and is not guaranteed.-- If Morphic can't connect to Ollama, verify it's running: `ollama list`-- Check that `OLLAMA_BASE_URL` points to the correct host (use `http://host.docker.internal:11434` when running Morphic in Docker)-- Review startup logs for model validation results+- **Validation Logs**: Check server logs on startup to verify your configured models:+ ```+ ✓ llama3.2:latest (configured and tools supported)+ ✗ codellama:latest (configured but lacks tools support)+ ```## Other Features