Added OpenAI and Fireworks vision

examples/foundational/12a-fireworks-describe-video.py

@@ -11,6 +11,7 @@
 from dailyai.pipeline.pipeline import Pipeline
 from dailyai.services.elevenlabs_ai_service import ElevenLabsTTSService
 from dailyai.services.open_ai_services import OpenAIVisionService
+from dailyai.services.fireworks_ai_services import FireworksVisionService
 from dailyai.transports.daily_transport import DailyTransport
 
 from runner import configure
@@ -62,8 +63,11 @@ async def main(room_url: str, token):
         vision_aggregator = VisionImageFrameAggregator()
 
         # If you run into weird description, try with use_cpu=True
-        img_desc = OpenAIVisionService(
-            api_key=os.getenv("OPENAI_API_KEY")
+        # img_desc = OpenAIVisionService(
+        #     api_key=os.getenv("OPENAI_API_KEY")
+        # )
+        img_desc = FireworksVisionService(
+            api_key=os.getenv("FIREWORKS_API_KEY")
         )
 
         tts = ElevenLabsTTSService(

src/dailyai/services/fireworks_ai_services.py

@@ -5,7 +5,7 @@
 
 from dailyai.services.ai_services import ImageGenService, VisionService
 from dailyai.services.openai_api_llm_service import BaseOpenAILLMService
-
+from dailyai.services.open_ai_services import OpenAIVisionService
 
 try:
     from openai import AsyncOpenAI
@@ -59,5 +59,7 @@ async def run_image_gen(self, prompt: str) -> tuple[str, bytes, tuple[int, int]]
             image = Image.open(image_stream)
             return (image_url, image.tobytes(), image.size)
 
-class FireworksVisionService(VisionService):
-
+
+class FireworksVisionService(OpenAIVisionService):
+    def __init__(self, *, api_key, model="accounts/fireworks/models/firellava-13b"):
+        super().__init__(model=model, api_key=api_key, base_url="https://api.fireworks.ai/inference/v1")

src/dailyai/services/open_ai_services.py

@@ -72,9 +72,14 @@ def __init__(
         *,
         model="gpt-4-vision-preview",
         api_key,
+        base_url=None,
     ):
         self._model = model
-        self._client = AsyncOpenAI(api_key=api_key)
+        if base_url:
+            self._client = AsyncOpenAI(api_key=api_key, base_url=base_url)
+        else:
+            self._client = AsyncOpenAI(api_key=api_key, base_url=base_url)
+
         super().__init__()
 
     async def run_vision_async(self, frame):