Feature/mira network integration

infernet_services/services/css_inference_service/.env.example

@@ -1,5 +1,6 @@
 OPENAI_API_KEY=
 GOOSEAI_API_KEY=
 PERPLEXITYAI_API_KEY=
+MIRANETWORK_API_KEY=
 
 RETRY_PARAMS={"tries": 3, "delay": 3, "backoff": 2, "max_delay": 10, "jitter": [0.5, 1.5]}

infernet_services/services/css_inference_service/src/app.py

@@ -95,6 +95,7 @@ def create_app() -> Quart:
         CSSProvider.GOOSEAI: os.getenv("GOOSEAI_API_KEY"),
         CSSProvider.OPENAI: os.getenv("OPENAI_API_KEY"),
         CSSProvider.PERPLEXITYAI: os.getenv("PERPLEXITYAI_API_KEY"),
+        CSSProvider.MIRANETWORK: os.getenv("MIRANETWORK_API_KEY"),
     }
     providers = sorted(api_keys.keys())
 

infernet_services/tests/css_inference_service/conftest.py

@@ -19,9 +19,10 @@
 log = logging.getLogger(__name__)
 
 env_vars = {
-    "PERPLEXITYAI_API_KEY": os.environ["PERPLEXITYAI_API_KEY"],
-    "GOOSEAI_API_KEY": os.environ["GOOSEAI_API_KEY"],
-    "OPENAI_API_KEY": os.environ["OPENAI_API_KEY"],
+    "PERPLEXITYAI_API_KEY": os.environ.get("PERPLEXITYAI_API_KEY", "dummy_key"),
+    "GOOSEAI_API_KEY": os.environ.get("GOOSEAI_API_KEY", "dummy_key"),
+    "OPENAI_API_KEY": os.environ.get("OPENAI_API_KEY", "dummy_key"),
+    "MIRANETWORK_API_KEY": os.environ.get("MIRANETWORK_API_KEY"),
     "CSS_INF_WORKFLOW_POSITIONAL_ARGS": "[]",
     "CSS_INF_WORKFLOW_KW_ARGS": json.dumps(
         {
@@ -51,7 +52,7 @@
         name=CSS_OPENAI_ONLY,
         image_id=f"ritualnetwork/{SERVICE_NAME}:{SERVICE_VERSION}",
         env_vars={
-            "OPENAI_API_KEY": os.environ["OPENAI_API_KEY"],
+            "OPENAI_API_KEY": os.environ.get("OPENAI_API_KEY", "dummy_key"),
             "CSS_INF_WORKFLOW_POSITIONAL_ARGS": "[]",
             "CSS_INF_WORKFLOW_KW_ARGS": json.dumps(
                 {

infernet_services/tests/css_inference_service/simple_mira_test.py

@@ -0,0 +1,65 @@
+"""
+Simple test for Mira Network integration.
+This script tests the Mira Network integration by making direct API calls.
+"""
+
+import unittest
+import os
+import sys
+from pathlib import Path
+import json
+import requests
+
+# Add the project root to the Python path
+project_root = str(Path(__file__).parent.parent.parent.parent)
+if project_root not in sys.path:
+    sys.path.append(project_root)
+
+# Use the API key from conftest.py
+MIRANETWORK_API_KEY = ""
+
+class TestMiraNetworkAPI(unittest.TestCase):
+    """Test class for Mira Network API integration"""
+
+    def test_mira_network_api_direct(self):
+        """Test Mira Network API directly without using the css_mux function"""
+        # Define the API endpoint
+        url = "https://api.mira.network/v1/chat/completions"
+
+        # Define the headers
+        headers = {
+            "Content-Type": "application/json",
+            "Authorization": f"Bearer {MIRANETWORK_API_KEY}"
+        }
+
+        # Define the payload
+        payload = {
+            "model": "gpt-4o",
+            "messages": [
+                {
+                    "role": "user",
+                    "content": "Explain quantum computing in one paragraph."
+                }
+            ],
+            "temperature": 0.7,
+            "max_tokens": 100
+        }
+
+        # Make the API call
+        response = requests.post(url, headers=headers, json=payload)
+
+        # Print the response for debugging
+        print(f"Status code: {response.status_code}")
+        print(f"Response: {response.text}")
+
+        # Verify the response
+        self.assertEqual(response.status_code, 200)
+        response_json = response.json()
+        self.assertIn("choices", response_json)
+        self.assertGreater(len(response_json["choices"]), 0)
+        self.assertIn("message", response_json["choices"][0])
+        self.assertIn("content", response_json["choices"][0]["message"])
+        self.assertGreater(len(response_json["choices"][0]["message"]["content"]), 0)
+
+if __name__ == "__main__":
+    unittest.main() 

libraries/infernet_ml/src/infernet_ml/utils/codec/css.py

@@ -26,7 +26,7 @@ class CSSProvider(IntEnum):
     OPENAI = 0
     GOOSEAI = 1
     PERPLEXITYAI = 2
-
+    MIRANETWORK = 3
 
 def encode_css_completion_request(
     provider: CSSProvider,

libraries/infernet_ml/src/infernet_ml/utils/css_models.py

@@ -85,4 +85,11 @@
         {"id": "GOOSEAI/gpt-neoNone-3b", "name": "GPT-Neo 1.3B", "parameters": "1.3B"},
         {"id": "GOOSEAI/gpt-neo-2-7b", "name": "GPT-Neo 2.7B", "parameters": "2.7B"},
     ],
+    CSSProvider.MIRANETWORK: [
+        {"id": "MIRANETWORK/gpt-4o", "name": "GPT-4O", "parameters": None},
+        {"id": "MIRANETWORK/deepseek-r1", "name": "Deepseek R1", "parameters": None},
+        {"id": "MIRANETWORK/gpt-4o-mini", "name": "GPT-4O Mini", "parameters": None},
+        {"id": "MIRANETWORK/claude-3.5-sonnet", "name": "Claude 3.5 Sonnet", "parameters": None},
+        {"id": "MIRANETWORK/llama-3.3-70b-instruct", "name": "Llama 3.3 70B Instruct", "parameters": "70B"},
+    ],
 }

libraries/infernet_ml/src/infernet_ml/utils/css_mux.py

@@ -1,7 +1,7 @@
 """
 Library containing functions for accessing closed source models.
 
-Currently, 3 APIs are supported: OPENAI, PERPLEXITYAI, and GOOSEAI.
+Currently, 4 APIs are supported: OPENAI, PERPLEXITYAI, GOOSEAI, and MIRANETWORK.
 
 """
 
@@ -51,7 +51,7 @@ class CSSProvider(StrEnum):
     OPENAI = "OPENAI"
     PERPLEXITYAI = "PERPLEXITYAI"
     GOOSEAI = "GOOSEAI"
-
+    MIRANETWORK = "MIRANETWORK"
 
 ApiKeys = Dict[CSSProvider, Optional[str]]
 
@@ -174,6 +174,29 @@ def goose_ai_request_generator(req: CSSRequest) -> tuple[str, dict[str, Any]]:
             raise InfernetMLException(f"Unsupported request {req}")
 
 
+def mira_network_request_generator(req: CSSRequest) -> tuple[str, dict[str, Any]]:
+    """Returns base url & json input for Miran Network API.
+
+    Args:
+        req: a CSSRequest object, containing provider, endpoint, model,
+        api keys & params.
+
+    Returns:
+        base_url: str
+        processed input: dict[str, Any]
+
+    Raises:
+        InfernetMLException: if an unsupported model or params specified.
+    """
+    match req:
+        case CSSRequest(model=model_name, params=CSSCompletionParams(messages=msgs)):
+            return "https://api.mira.network/v1/", {
+                "model": model_name,
+                "messages": [msg.model_dump() for msg in msgs],
+            }
+        case _:
+            raise InfernetMLException(f"Unsupported request {req}")
+
 def extract_completions(result: Dict[str, Any]) -> str:
     return cast(str, result["choices"][0]["message"]["content"])
 
@@ -214,6 +237,15 @@ def extract_completions_gooseai(result: Dict[str, Any]) -> str:
             }
         },
     },
+    CSSProvider.MIRANETWORK: {
+        "input_func": mira_network_request_generator,
+        "endpoints": {
+            "completions": {
+                "real_endpoint": "chat/completions",
+                "post_process": extract_completions,
+            }
+        },
+    },
 }
 
 
@@ -289,6 +321,9 @@ def css_mux(req: CSSRequest) -> str:
             case CSSProvider.PERPLEXITYAI:
                 if result.status_code == 429:
                     raise RetryableException(result.text)
+            case CSSProvider.MIRANETWORK:
+                if result.status_code == 429 or result.status_code == 500:
+                    raise RetryableException(result.text)
             case _:
                 raise InfernetMLException(result.text)
 
@@ -304,6 +339,7 @@ def css_mux(req: CSSRequest) -> str:
         "content", ""
     ),
     CSSProvider.GOOSEAI: lambda result: result["choices"][0]["text"],
+    CSSProvider.MIRANETWORK: lambda result: result["choices"][0]["delta"].get("content", ""),
 }