update-endpoints

docs/issues.md

@@ -1,6 +1 @@
-2025-03-17 22:33:24,340 - parler_tts.modeling_parler_tts - WARNING - `prompt_attention_mask` is specified but `attention_mask` is not. A full `attention_mask` will be created. Make sure this is the intended behaviour.
-W0317 22:33:36.322000 1 torch/_inductor/utils.py:1137] [0/0] Not enough SMs to use max_autotune_gemm mode
-CUDAGraph supports dynamic shapes by recording a new graph for each distinct input size. Recording too many CUDAGraphs may lead to extra overhead. We have observed 51 distinct sizes. Please consider the following options for better performance: a) padding inputs to a few fixed number of shapes; or b) set torch._inductor.config.triton.cudagraph_skip_dynamic_graphs=True. Set torch._inductor.config.triton.cudagraph_dynamic_shape_warn_limit=None to silence this warning.
- 
-
- 
+https://github.com/codelion/optillm

src/server/main.py

@@ -579,7 +579,7 @@ limiter = Limiter(key_func=get_remote_address)
 app.state.limiter = limiter
 
 # Endpoints
-@app.post("/audio/speech", response_class=StreamingResponse)
+@app.post("/v1/audio/speech", response_class=StreamingResponse)
 async def synthesize_kannada(request: KannadaSynthesizeRequest):
     if not tts_manager.model:
         raise HTTPException(status_code=503, detail="TTS model not loaded")
@@ -593,7 +593,7 @@ async def synthesize_kannada(request: KannadaSynthesizeRequest):
         headers={"Content-Disposition": "attachment; filename=synthesized_kannada_speech.wav"}
     )
 
-@app.post("/translate", response_model=TranslationResponse)
+@app.post("/v1/translate", response_model=TranslationResponse)
 async def translate(request: TranslationRequest, translate_manager: TranslateManager = Depends(get_translate_manager)):
     if not request.sentences:
         raise HTTPException(status_code=400, detail="Input sentences are required")
@@ -827,7 +827,7 @@ async def chat_v2(
             logger.error(f"Error processing request: {str(e)}")
             raise HTTPException(status_code=500, detail=f"An error occurred: {str(e)}")
 
-@app.post("/transcribe/", response_model=TranscriptionResponse)
+@app.post("/v1/transcribe/", response_model=TranscriptionResponse)
 async def transcribe_audio(file: UploadFile = File(...), language: str = Query(..., enum=list(asr_manager.model_language.keys()))):
     async with request_queue:
         if not asr_manager.model: