infernet-1.0.0 update

projects/gpt4/container/src/app.py

@@ -1,8 +1,16 @@
 import logging
+import os
 from typing import Any, cast
 
 from eth_abi import decode, encode  # type: ignore
-from infernet_ml.utils.service_models import InfernetInput, InfernetInputSource
+from infernet_ml.utils.css_mux import (
+    ConvoMessage,
+    CSSCompletionParams,
+    CSSRequest,
+    Provider,
+)
+from infernet_ml.utils.service_models import InfernetInput
+from infernet_ml.utils.service_models import JobLocation
 from infernet_ml.workflows.inference.css_inference_workflow import CSSInferenceWorkflow
 from quart import Quart, request
 
@@ -12,7 +20,9 @@ log = logging.getLogger(__name__)
 def create_app() -> Quart:
     app = Quart(__name__)
 
-    workflow = CSSInferenceWorkflow(provider="OPENAI", endpoint="completions")
+    workflow = CSSInferenceWorkflow(
+        api_keys={Provider.OPENAI: os.environ["OPENAI_API_KEY"]}
+    )
 
     workflow.setup()
 
@@ -24,7 +34,7 @@ def create_app() -> Quart:
         return "GPT4 Example Program"
 
     @app.route("/service_output", methods=["POST"])
-    async def inference() -> dict[str, Any]:
+    async def inference() -> Any:
         req_data = await request.get_json()
         """
         InfernetInput has the format:
@@ -33,52 +43,62 @@ def create_app() -> Quart:
         """
         infernet_input: InfernetInput = InfernetInput(**req_data)
 
-        if infernet_input.source == InfernetInputSource.OFFCHAIN:
-            prompt = cast(dict[str, Any], infernet_input.data).get("prompt")
-        else:
-            # On-chain requests are sent as a generalized hex-string which we will
-            # decode to the appropriate format.
-            (prompt,) = decode(
-                ["string"], bytes.fromhex(cast(str, infernet_input.data))
-            )
+        match infernet_input:
+            case InfernetInput(source=JobLocation.OFFCHAIN):
+                prompt = cast(dict[str, Any], infernet_input.data).get("prompt")
+            case InfernetInput(source=JobLocation.ONCHAIN):
+                # On-chain requests are sent as a generalized hex-string which we will
+                # decode to the appropriate format.
+                (prompt,) = decode(
+                    ["string"], bytes.fromhex(cast(str, infernet_input.data))
+                )
+            case _:
+                raise ValueError("Invalid source")
 
-        result: dict[str, Any] = workflow.inference(
-            {
-                "model": "gpt-4-0613",
-                "params": {
-                    "endpoint": "completions",
-                    "messages": [
-                        {"role": "system", "content": "You are a helpful assistant."},
-                        {"role": "user", "content": prompt},
-                    ],
-                },
-            }
+        result = workflow.inference(
+            CSSRequest(
+                provider=Provider.OPENAI,
+                endpoint="completions",
+                model="gpt-4-0613",
+                params=CSSCompletionParams(
+                    messages=[
+                        ConvoMessage(
+                            role="system", content="you are a helpful " "assistant."
+                        ),
+                        ConvoMessage(role="user", content=cast(str, prompt)),
+                    ]
+                ),
+            )
         )
 
-        if infernet_input.source == InfernetInputSource.OFFCHAIN:
-            """
-            In case of an off-chain request, the result is returned as is.
-            """
-            return {"message": result}
-        else:
-            """
-            In case of an on-chain request, the result is returned in the format:
-            {
-                "raw_input": str,
-                "processed_input": str,
-                "raw_output": str,
-                "processed_output": str,
-                "proof": str,
-            }
-            refer to: https://docs.ritual.net/infernet/node/containers for more info.
-            """
-            return {
-                "raw_input": "",
-                "processed_input": "",
-                "raw_output": encode(["string"], [result]).hex(),
-                "processed_output": "",
-                "proof": "",
-            }
+        match infernet_input:
+            case InfernetInput(destination=JobLocation.OFFCHAIN):
+                """
+                In case of an off-chain request, the result is returned as is.
+                """
+                return {"message": result}
+            case InfernetInput(destination=JobLocation.ONCHAIN):
+                """
+                In case of an on-chain request, the result is returned in the format:
+                {
+                    "raw_input": str,
+                    "processed_input": str,
+                    "raw_output": str,
+                    "processed_output": str,
+                    "proof": str,
+                }
+                refer to: https://docs.ritual.net/infernet/node/containers for more
+                info.
+                """
+                return {
+                    "raw_input": "",
+                    "processed_input": "",
+                    "raw_output": encode(["string"], [result]).hex(),
+                    "processed_output": "",
+                    "proof": "",
+                }
+            case _:
+                raise ValueError("Invalid destination")
 
     return app
 

projects/gpt4/container/src/requirements.txt

@@ -1,5 +1,4 @@
 quart==0.19.4
-infernet_ml==0.1.0
-PyArweave @ git+https://github.com/ritual-net/pyarweave.git
+infernet-ml==1.0.0
+infernet-ml[css_inference]==1.0.0
 web3==6.15.0
-retry2==0.9.5