local support for Arch-Router via Ollama (#509)

* local support for Arch-Router via Ollama

* fixed issue withe non-local yaml config

---------

Co-authored-by: Salman Paracha <salmanparacha@MacBook-Pro-329.local>

demos/use_cases/preference_based_routing/arch_config.yaml

@@ -30,7 +30,7 @@ llm_providers:
 
   - name: code_generation
     access_key: $OPENAI_API_KEY
-    provider_interface: claude
+    provider_interface: openai
     model: gpt-4.1
     usage: generating new code snippets, functions, or boilerplate based on user prompts or requirements
 

demos/use_cases/preference_based_routing/arch_config_local.yaml

@@ -0,0 +1,45 @@
+version: v0.1.0
+
+routing:
+  model: arch-router
+
+listeners:
+  egress_traffic:
+    address: 0.0.0.0
+    port: 12000
+    message_format: openai
+    timeout: 30s
+
+llm_providers:
+
+  - name: arch-router
+    access_key: $ARCH_API_KEY
+    provider_interface: arch
+    model: hf.co/katanemo/Arch-Router-1.5B.gguf:Q4_K_M
+    endpoint: host.docker.internal:11434
+
+  - name: gpt-4o-mini
+    provider_interface: openai
+    access_key: $OPENAI_API_KEY
+    model: gpt-4o-mini
+
+  - name: gpt-4.1
+    provider_interface: openai
+    access_key: $OPENAI_API_KEY
+    model: gpt-4.1
+    default: true
+
+  - name: code_generation
+    access_key: $OPENAI_API_KEY
+    provider_interface: openai
+    model: gpt-4.1
+    usage: generating new code snippets, functions, or boilerplate based on user prompts or requirements
+
+  - name: code_understanding
+    provider_interface: openai
+    access_key: $OPENAI_API_KEY
+    model: gpt-4.1
+    usage: understand and explain existing code snippets, functions, or libraries
+
+tracing:
+  random_sampling: 100