Adding gpt-4o-mini config for seamless experimentation

config/models.yaml

@@ -41,6 +41,15 @@ gpt-4o:
     max_tokens: 500
     temperature: 1.0
 
+gpt-4o-mini:
+  model_type: openai
+  model: gpt-4o-mini
+  # URL and api_key should be defined at .env file as GRASP_GPT-4O-MINI_URL and GRASP_GPT-4O-MINI_TOKEN
+  api_version: 2024-08-01-preview
+  parameters:
+    max_tokens: 5000
+    temperature: 0.0001
+
 #QWEN VL 72b deployed in vllm
 qwen_vl_72b:
   # URL and api_key should be defined at .env file as GRASP_QWEN_VL_72B_URL and GRASP_QWEN_VL_72B_TOKEN

tasks/examples/structured_output_with_multi_llm/dpo_samples/README.md

@@ -15,7 +15,7 @@ The DPO Samples task follows this workflow:
    - Initialize state variables for tracking ratings
 
 2. **Generation Phase**:
-   - Send the user prompt to multiple LLMs (gpt4, gpt-4o, gpt4o-mini)
+   - Send the user prompt to multiple LLMs (gpt4, gpt-4o, gpt-4o-mini)
    - Collect structured responses from each model
 
 3. **Rating Phase**:
@@ -47,7 +47,7 @@ The task uses three custom data transforms:
 The graph consists of two main nodes:
 
 1. **generate_samples (multi_llm)**:
-   - Uses three different models: gpt4, gpt-4o, and gpt4o-mini
+   - Uses three different models: gpt4, gpt-4o, and gpt-4o-mini
    - Each model has different structured output configurations:
      - gpt4: YAML-defined schema
      ```yaml
@@ -68,7 +68,7 @@ The graph consists of two main nodes:
       schema: "models.structured_output.schemas_factory.SimpleResponse"
 
      ```
-     - gpt4o-mini: YAML-defined schema with structured output disabled
+     - gpt-4o-mini: YAML-defined schema with structured output disabled
      ```yaml
      structured_output:
       enabled: false
@@ -129,7 +129,7 @@ To run the task:
 ```json
 {
   "id": "test_id",
-  "annotation_type": ["scale", "gpt4", "gpt-4o", "gpt4o-mini"],
+  "annotation_type": ["scale", "gpt4", "gpt-4o", "gpt-4o-mini"],
   "language": "en",
   "tags": ["dpo_samples_rating"],
   "conversation": [
@@ -154,7 +154,7 @@ To run the task:
 {
   "id": "test_id",
   "taxonomy": ["test_taxonomy"],
-  "annotation_type": ["scale", "gpt4", "gpt-4o", "gpt4o-mini"],
+  "annotation_type": ["scale", "gpt4", "gpt-4o", "gpt-4o-mini"],
   "language": "en",
   "tags": ["dpo_samples_rating"],
   "conversation": [
@@ -185,7 +185,7 @@ To run the task:
         },
         {
           "generation": "A sustainable urban transportation system should focus on...",
-          "model": "gpt4o-mini",
+          "model": "gpt-4o-mini",
           "judge_rating": 6,
           "judge_explanation": "The response covers basic aspects but lacks depth..."
         }

tasks/examples/structured_output_with_multi_llm/dpo_samples/graph_config.yaml

@@ -41,8 +41,8 @@ graph_config:
             max_tokens: 2000
           structured_output:
             schema: "core.models.structured_output.schemas_factory.SimpleResponse"
-        gpt4o-mini:
-          name: gpt4o-mini
+        gpt-4o-mini:
+          name: gpt-4o-mini
           parameters:
             temperature: 0.7
             max_tokens: 2000
@@ -102,7 +102,7 @@ output_config:
         "scale",
         "gpt4",
         "gpt-4o",
-        "gpt4o-mini"
+        "gpt-4o-mini"
       ]
     language:
       value: "en"