1. load configure from environment 2. support anthropic base url

backend/config.py

@@ -3,10 +3,13 @@
 # TODO: Should only be set to true when value is 'True', not any abitrary truthy value
 import os
 
-ANTHROPIC_API_KEY = os.environ.get("ANTHROPIC_API_KEY", None)
+CFG_ANTHROPIC_API_KEY = os.environ.get("ANTHROPIC_API_KEY", None)
+CFG_ANTHROPIC_BASE_URL = os.environ.get("ANTHROPIC_BASE_URL", None)
+
+CFG_OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY",None)
+CFG_OPENAI_BASE_URL = os.environ.get("OPENAI_BASE_URL",None)
 
 # Debugging-related
-
 SHOULD_MOCK_AI_RESPONSE = bool(os.environ.get("MOCK", False))
 IS_DEBUG_ENABLED = bool(os.environ.get("IS_DEBUG_ENABLED", False))
 DEBUG_DIR = os.environ.get("DEBUG_DIR", "")

backend/evals/core.py

@@ -1,5 +1,5 @@
 import os
-from config import ANTHROPIC_API_KEY
+from config import CFG_ANTHROPIC_API_KEY,CFG_ANTHROPIC_BASE_URL,CFG_OPENAI_API_KEY,CFG_OPENAI_BASE_URL
 
 from llm import Llm, stream_claude_response, stream_openai_response
 from prompts import assemble_prompt
@@ -8,30 +8,28 @@ from prompts.types import Stack
 
 async def generate_code_core(image_url: str, stack: Stack, model: Llm) -> str:
     prompt_messages = assemble_prompt(image_url, stack)
-    openai_api_key = os.environ.get("OPENAI_API_KEY")
-    anthropic_api_key = ANTHROPIC_API_KEY
-    openai_base_url = None
 
     async def process_chunk(content: str):
         pass
 
     if model == Llm.CLAUDE_3_SONNET:
-        if not anthropic_api_key:
+        if not CFG_ANTHROPIC_API_KEY:
             raise Exception("Anthropic API key not found")
 
         completion = await stream_claude_response(
             prompt_messages,
-            api_key=anthropic_api_key,
+            api_key=CFG_ANTHROPIC_API_KEY,
             callback=lambda x: process_chunk(x),
+            base_url=CFG_ANTHROPIC_BASE_URL
         )
     else:
-        if not openai_api_key:
+        if not CFG_OPENAI_API_KEY:
             raise Exception("OpenAI API key not found")
 
         completion = await stream_openai_response(
             prompt_messages,
-            api_key=openai_api_key,
+            api_key=CFG_OPENAI_API_KEY,
-            base_url=openai_base_url,
+            base_url=CFG_OPENAI_BASE_URL,
             callback=lambda x: process_chunk(x),
             model=model,
         )

backend/llm.py

@@ -73,9 +73,10 @@ async def stream_claude_response(
     messages: List[ChatCompletionMessageParam],
     api_key: str,
     callback: Callable[[str], Awaitable[None]],
+    base_url:str|None = None
 ) -> str:
 
-    client = AsyncAnthropic(api_key=api_key)
+    client = AsyncAnthropic(api_key=api_key,base_url=base_url)
 
     # Base parameters
     model = Llm.CLAUDE_3_SONNET
@@ -135,9 +136,10 @@ async def stream_claude_response_native(
     callback: Callable[[str], Awaitable[None]],
     include_thinking: bool = False,
     model: Llm = Llm.CLAUDE_3_OPUS,
+    base_url: str | None = None,
 ) -> str:
 
-    client = AsyncAnthropic(api_key=api_key)
+    client = AsyncAnthropic(api_key=api_key,base_url=base_url)
 
     # Base model parameters
     max_tokens = 4096

backend/routes/generate_code.py

@@ -2,7 +2,7 @@ import os
 import traceback
 from fastapi import APIRouter, WebSocket
 import openai
-from config import ANTHROPIC_API_KEY, IS_PROD, SHOULD_MOCK_AI_RESPONSE
+from config import CFG_ANTHROPIC_API_KEY,CFG_ANTHROPIC_BASE_URL,CFG_OPENAI_API_KEY,CFG_OPENAI_BASE_URL, IS_PROD, SHOULD_MOCK_AI_RESPONSE
 from custom_types import InputMode
 from llm import (
     Llm,
@@ -105,7 +105,7 @@ async def stream_code(websocket: WebSocket):
         openai_api_key = params["openAiApiKey"]
         print("Using OpenAI API key from client-side settings dialog")
     else:
-        openai_api_key = os.environ.get("OPENAI_API_KEY")
+        openai_api_key = CFG_OPENAI_API_KEY
         if openai_api_key:
             print("Using OpenAI API key from environment variable")
 
@@ -119,7 +119,7 @@ async def stream_code(websocket: WebSocket):
             "No OpenAI API key found. Please add your API key in the settings dialog or add it to backend/.env file. If you add it to .env, make sure to restart the backend server."
         )
         return
-
+    openai_api_key:str = openai_api_key
     # Get the OpenAI Base URL from the request. Fall back to environment variable if not provided.
     openai_base_url = None
     # Disable user-specified OpenAI Base URL in prod
@@ -219,7 +219,7 @@ async def stream_code(websocket: WebSocket):
     else:
         try:
             if validated_input_mode == "video":
-                if not ANTHROPIC_API_KEY:
+                if not CFG_ANTHROPIC_API_KEY:
                     await throw_error(
                         "Video only works with Anthropic models. No Anthropic API key found. Please add the environment variable ANTHROPIC_API_KEY to backend/.env"
                     )
@@ -228,14 +228,15 @@ async def stream_code(websocket: WebSocket):
                 completion = await stream_claude_response_native(
                     system_prompt=VIDEO_PROMPT,
                     messages=prompt_messages,  # type: ignore
-                    api_key=ANTHROPIC_API_KEY,
+                    api_key=CFG_ANTHROPIC_API_KEY,
                     callback=lambda x: process_chunk(x),
                     model=Llm.CLAUDE_3_OPUS,
                     include_thinking=True,
+                    base_url=CFG_ANTHROPIC_BASE_URL
                 )
                 exact_llm_version = Llm.CLAUDE_3_OPUS
             elif code_generation_model == Llm.CLAUDE_3_SONNET:
-                if not ANTHROPIC_API_KEY:
+                if not CFG_ANTHROPIC_API_KEY:
                     await throw_error(
                         "No Anthropic API key found. Please add the environment variable ANTHROPIC_API_KEY to backend/.env"
                     )
@@ -243,17 +244,19 @@ async def stream_code(websocket: WebSocket):
 
                 completion = await stream_claude_response(
                     prompt_messages,  # type: ignore
-                    api_key=ANTHROPIC_API_KEY,
+                    api_key=CFG_ANTHROPIC_API_KEY,
                     callback=lambda x: process_chunk(x),
+                    base_url=CFG_ANTHROPIC_BASE_URL
                 )
                 exact_llm_version = code_generation_model
             else:
                 completion = await stream_openai_response(
                     prompt_messages,  # type: ignore
                     api_key=openai_api_key,
-                    base_url=openai_base_url,
+                    base_url=CFG_OPENAI_BASE_URL,
                     callback=lambda x: process_chunk(x),
                     model=code_generation_model,
+
                 )
                 exact_llm_version = code_generation_model
         except openai.AuthenticationError as e:
@@ -307,7 +310,7 @@ async def stream_code(websocket: WebSocket):
             updated_html = await generate_images(
                 completion,
                 api_key=openai_api_key,
-                base_url=openai_base_url,
+                base_url=CFG_OPENAI_BASE_URL,
                 image_cache=image_cache,
             )
         else:

backend/video_to_app.py

@@ -14,7 +14,7 @@ import asyncio
 from datetime import datetime
 from prompts.claude_prompts import VIDEO_PROMPT
 from utils import pprint_prompt
-from config import ANTHROPIC_API_KEY
+from config import CFG_ANTHROPIC_API_KEY,CFG_ANTHROPIC_BASE_URL,CFG_OPENAI_API_KEY,CFG_OPENAI_BASE_URL
 from video.utils import extract_tag_content, assemble_claude_prompt_video
 from llm import (
     Llm,
@@ -32,7 +32,7 @@ async def main():
     video_filename = "shortest.mov"
     is_followup = False
 
-    if not ANTHROPIC_API_KEY:
+    if not CFG_ANTHROPIC_API_KEY:
         raise ValueError("ANTHROPIC_API_KEY is not set")
 
     # Get previous HTML
@@ -84,10 +84,11 @@ async def main():
     completion = await stream_claude_response_native(
         system_prompt=VIDEO_PROMPT,
         messages=prompt_messages,
-        api_key=ANTHROPIC_API_KEY,
+        api_key=CFG_ANTHROPIC_API_KEY,
         callback=lambda x: process_chunk(x),
         model=Llm.CLAUDE_3_OPUS,
         include_thinking=True,
+        base_url=CFG_ANTHROPIC_BASE_URL
     )
 
     end_time = time.time()