Update docs

docker/README.md

@@ -25,9 +25,10 @@ Create a configuration file `haiku.rag.yaml`:
 environment: production
 
 embeddings:
-  provider: ollama
+  model:
-  model: nomic-embed-text
+    provider: ollama
-  vector_dim: 768
+    name: nomic-embed-text
+    vector_dim: 768
 
 qa:
   model:

docs/configuration/index.md

@@ -35,9 +35,10 @@ A minimal configuration file with defaults:
 environment: production
 
 embeddings:
-  provider: ollama
+  model:
-  model: qwen3-embedding:4b
+    provider: ollama
-  vector_dim: 2560
+    name: qwen3-embedding:4b
+    vector_dim: 2560
 
 qa:
   model:
@@ -69,9 +70,10 @@ lancedb:
   region: ""
 
 embeddings:
-  provider: ollama
+  model:
-  model: qwen3-embedding:4b
+    provider: ollama
-  vector_dim: 2560
+    name: qwen3-embedding:4b
+    vector_dim: 2560
 
 reranking:
   model:
@@ -149,7 +151,7 @@ When using haiku.rag as a Python library, you can pass configuration directly to
 
 ```python
 from haiku.rag.config import AppConfig
-from haiku.rag.config.models import ModelConfig, QAConfig, EmbeddingsConfig
+from haiku.rag.config.models import EmbeddingModelConfig, ModelConfig, QAConfig, EmbeddingsConfig
 from haiku.rag.client import HaikuRAG
 
 # Create custom configuration
@@ -157,16 +159,16 @@ custom_config = AppConfig(
     qa=QAConfig(
         model=ModelConfig(
             provider="openai",
-            model="gpt-4o",
+            name="gpt-4o",
             temperature=0.7
         )
     ),
     embeddings=EmbeddingsConfig(
-        model=ModelConfig(
+        model=EmbeddingModelConfig(
             provider="ollama",
-            model="qwen3-embedding:4b"
+            name="qwen3-embedding:4b",
-        ),
+            vector_dim=2560
-        vector_dim=2560
+        )
     ),
     processing={"chunk_size": 512}
 )

docs/configuration/providers.md

@@ -73,9 +73,10 @@ If you use Ollama, you can use any pulled model that supports embeddings.
 
 ```yaml
 embeddings:
-  provider: ollama
+  model:
-  model: mxbai-embed-large
+    provider: ollama
-  vector_dim: 1024
+    name: mxbai-embed-large
+    vector_dim: 1024
 ```
 
 The Ollama base URL can be configured in your config file or via environment variable:
@@ -104,9 +105,10 @@ uv pip install haiku.rag-slim[voyageai]
 
 ```yaml
 embeddings:
-  provider: voyageai
+  model:
-  model: voyage-3.5
+    provider: voyageai
-  vector_dim: 1024
+    name: voyage-3.5
+    vector_dim: 1024
 ```
 
 Set your API key via environment variable:
@@ -121,9 +123,10 @@ OpenAI embeddings are included in the default installation:
 
 ```yaml
 embeddings:
-  provider: openai
+  model:
-  model: text-embedding-3-small  # or text-embedding-3-large
+    provider: openai
-  vector_dim: 1536
+    name: text-embedding-3-small  # or text-embedding-3-large
+    vector_dim: 1536
 ```
 
 Set your API key via environment variable:
@@ -138,9 +141,10 @@ For high-performance local inference, you can use vLLM to serve embedding models
 
 ```yaml
 embeddings:
-  provider: vllm
+  model:
-  model: mixedbread-ai/mxbai-embed-large-v1
+    provider: vllm
-  vector_dim: 512
+    name: mixedbread-ai/mxbai-embed-large-v1
+    vector_dim: 512
 
 providers:
   vllm:
@@ -155,9 +159,10 @@ providers:
 
 ```yaml
 embeddings:
-  provider: lm_studio
+  model:
-  model: text-embedding-qwen3-embedding-4b
+    provider: lm_studio
-  vector_dim: 2560
+    name: text-embedding-qwen3-embedding-4b
+    vector_dim: 2560
 
 providers:
   lm_studio:

docs/tutorial.md

@@ -33,9 +33,10 @@ Configure haiku.rag to use OpenAI. Create a `haiku.rag.yaml` file:
 
 ```yaml
 embeddings:
-  provider: openai
+  model:
-  model: text-embedding-3-small  # or text-embedding-3-large
+    provider: openai
-  vector_dim: 1536
+    name: text-embedding-3-small  # or text-embedding-3-large
+    vector_dim: 1536
 
 qa:
   model: