Update models

src/aichat/config.yaml

@@ -3,7 +3,7 @@
 rag_embedding_model: ollama:nomic-embed-text
 rag_chunk_size: 8192
 rag_chunk_overlap: 409
-model: ollama:qwen2.5
+model: ollama:gemma3:12b-it-qat
 temperature: 0
 clients:
 - type: openai-compatible
@@ -11,27 +11,21 @@ clients:
   api_base: http://localhost:11434/v1
   api_key: __LLM_API_KEY__
   models:
-    - name: qwen2.5
+    - name: gemma3:4b
       supports_function_calling: true
       max_input_tokens: 128000
-    - name: qwen2.5-coder:32b
+    - name: gemma3:12b-it-qat
       supports_function_calling: true
       max_input_tokens: 128000
-    - name: nomic-embed-text
+    - name: gemma3:27b-it-qat
-      type: embedding
-      max_input_tokens: 200000
-      max_tokens_per_chunk: 2000
-      default_chunk_size: 8192
-      max_batch_size: 100
-    - name: gemma2
       supports_function_calling: true
       max_input_tokens: 128000
-    - name: mistral
+    - name: gpt-oss:20b
       supports_function_calling: true
       max_input_tokens: 128000
-    - name: deepseek-r1:7b
+    - name: llama3.1:8b
       supports_function_calling: true
       max_input_tokens: 128000
-    - name: deepcoder:14b
+    - name: qwen2.5:7b
       supports_function_calling: true
       max_input_tokens: 128000