isa-model 0.3.5__py3-none-any.whl → 0.3.6__py3-none-any.whl

isa_model/serving/api/routes/unified.py

@@ -0,0 +1,202 @@
+"""
+Unified API Route - Single endpoint for all AI services
+
+This is the main API that handles all types of AI requests:
+- Vision tasks (image analysis, OCR, UI detection)
+- Text tasks (chat, generation, translation) 
+- Audio tasks (TTS, STT)
+- Image generation tasks
+- Embedding tasks
+"""
+
+from fastapi import APIRouter, HTTPException, UploadFile, File, Form
+from pydantic import BaseModel, Field
+from typing import Optional, Dict, Any, Union, List
+import logging
+import asyncio
+from pathlib import Path
+
+from isa_model.client import ISAModelClient
+
+logger = logging.getLogger(__name__)
+router = APIRouter()
+
+class UnifiedRequest(BaseModel):
+    """Unified request model for all AI services"""
+    input_data: Union[str, Dict[str, Any]] = Field(..., description="Input data (text, image URL, etc.)")
+    task: str = Field(..., description="Task to perform (chat, analyze_image, generate_speech, etc.)")
+    service_type: str = Field(..., description="Service type (text, vision, audio, image, embedding)")
+    model_hint: Optional[str] = Field(None, description="Optional model preference")
+    provider_hint: Optional[str] = Field(None, description="Optional provider preference")
+    parameters: Optional[Dict[str, Any]] = Field(default_factory=dict, description="Additional task parameters")
+
+class UnifiedResponse(BaseModel):
+    """Unified response model for all AI services"""
+    success: bool
+    result: Optional[Any] = None
+    error: Optional[str] = None
+    metadata: Dict[str, Any]
+
+# Global ISA client instance for server-side processing
+_isa_client = None
+
+def get_isa_client():
+    """Get or create ISA client for local processing"""
+    global _isa_client
+    if _isa_client is None:
+        _isa_client = ISAModelClient(mode="local")  # Use local mode
+    return _isa_client
+
+@router.get("/")
+async def unified_info():
+    """API information"""
+    return {
+        "service": "unified_api",
+        "status": "active",
+        "description": "Single endpoint for all AI services",
+        "supported_service_types": ["vision", "text", "audio", "image", "embedding"],
+        "version": "1.0.0"
+    }
+
+@router.post("/invoke", response_model=UnifiedResponse)
+async def unified_invoke(request: UnifiedRequest) -> UnifiedResponse:
+    """
+    **Unified API endpoint for all AI services**
+    
+    This single endpoint handles:
+    - Vision: image analysis, OCR, UI detection
+    - Text: chat, generation, translation
+    - Audio: TTS, STT, transcription
+    - Image: generation, img2img
+    - Embedding: text embedding, similarity
+    
+    **Uses ISAModelClient in local mode - all the complex logic is in client.py**
+    """
+    try:
+        # Get ISA client instance (local mode)
+        client = get_isa_client()
+        
+        # Use client's local invoke method directly
+        # This handles all the complexity: model selection, service routing, execution
+        result = await client._invoke_local(
+            input_data=request.input_data,
+            task=request.task,
+            service_type=request.service_type,
+            model_hint=request.model_hint,
+            provider_hint=request.provider_hint,
+            **request.parameters
+        )
+        
+        # Return the result in our API format
+        return UnifiedResponse(
+            success=result["success"],
+            result=result.get("result"),
+            error=result.get("error"),
+            metadata=result["metadata"]
+        )
+        
+    except Exception as e:
+        logger.error(f"Unified invoke failed: {e}")
+        return UnifiedResponse(
+            success=False,
+            error=str(e),
+            metadata={
+                "task": request.task,
+                "service_type": request.service_type,
+                "model_hint": request.model_hint,
+                "provider_hint": request.provider_hint
+            }
+        )
+
+@router.post("/invoke-file", response_model=UnifiedResponse)
+async def unified_invoke_file(
+    task: str = Form(...),
+    service_type: str = Form(...),
+    model_hint: Optional[str] = Form(None),
+    provider_hint: Optional[str] = Form(None),
+    file: UploadFile = File(...)
+) -> UnifiedResponse:
+    """
+    Unified file upload endpoint
+    
+    For tasks that require file input (images, audio, documents)
+    """
+    try:
+        # Read file data
+        file_data = await file.read()
+        
+        # Get ISA client instance (local mode)
+        client = get_isa_client()
+        
+        # Use client's local invoke method with binary data
+        result = await client._invoke_local(
+            input_data=file_data,  # Binary data
+            task=task,
+            service_type=service_type,
+            model_hint=model_hint,
+            provider_hint=provider_hint,
+            filename=file.filename,
+            content_type=file.content_type,
+            file_size=len(file_data)
+        )
+        
+        # Return the result in our API format
+        return UnifiedResponse(
+            success=result["success"],
+            result=result.get("result"),
+            error=result.get("error"),
+            metadata={
+                **result["metadata"],
+                "filename": file.filename,
+                "content_type": file.content_type,
+                "file_size": len(file_data)
+            }
+        )
+        
+    except Exception as e:
+        logger.error(f"File invoke failed: {e}")
+        return UnifiedResponse(
+            success=False,
+            error=str(e),
+            metadata={
+                "task": task,
+                "service_type": service_type,
+                "filename": file.filename if file else None
+            }
+        )
+
+@router.get("/models")
+async def get_available_models(service_type: Optional[str] = None):
+    """Get available models (optional filter by service type)"""
+    try:
+        client = get_isa_client()
+        return await client.get_available_models(service_type)
+    except Exception as e:
+        logger.error(f"Failed to get available models: {e}")
+        # Fallback static model list
+        return {
+            "models": [
+                {"service_type": "vision", "provider": "openai", "model_id": "gpt-4.1-mini"},
+                {"service_type": "text", "provider": "openai", "model_id": "gpt-4.1-mini"},
+                {"service_type": "audio", "provider": "openai", "model_id": "whisper-1"},
+                {"service_type": "audio", "provider": "openai", "model_id": "tts-1"},
+                {"service_type": "embedding", "provider": "openai", "model_id": "text-embedding-3-small"},
+                {"service_type": "image", "provider": "replicate", "model_id": "black-forest-labs/flux-schnell"}
+            ]
+        }
+
+@router.get("/health")
+async def health_check():
+    """Health check for unified API"""
+    try:
+        client = get_isa_client()
+        health_result = await client.health_check()
+        return {
+            "api": "healthy",
+            "client_health": health_result
+        }
+    except Exception as e:
+        return {
+            "api": "error",
+            "error": str(e)
+        }

{isa_model-0.3.5.dist-info → isa_model-0.3.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: isa_model
-Version: 0.3.5
+Version: 0.3.6
 Summary: Unified AI model serving framework
 Author: isA_Model Contributors
 Classifier: Development Status :: 3 - Alpha
@@ -34,6 +34,9 @@ Requires-Dist: accelerate>=0.20.0
 Requires-Dist: bitsandbytes>=0.39.0
 Requires-Dist: peft>=0.4.0
 Requires-Dist: trl>=0.4.0
+Requires-Dist: supabase>=2.0.0
+Requires-Dist: pgvector>=0.2.0
+Requires-Dist: psycopg2-binary>=2.9.0
 
 # isa_model_sdk - Unified AI Model Serving Framework
 

{isa_model-0.3.5.dist-info → isa_model-0.3.6.dist-info}/RECORD

@@ -1,19 +1,26 @@
-isa_model/__init__.py,sha256=skxx7AA-1BzIT_UaDHcNmIo4rEhgL8MqOk8vPpZPrAo,87
-isa_model/config/__init__.py,sha256=FHWgjehhr-Mk_qDj7ZqGYLP1Iwc5F7aT93Owk_lymTk,262
-isa_model/config/config_manager.py,sha256=jedSOySXpyYw-xA1pKP6w1fAdpFSoAgzrNNRKYz4mgU,6720
-isa_model/core/model_manager.py,sha256=-lffVoNdy19pDRoag9v-jYfppcY-hua0ciPl5_BLTw4,8732
-isa_model/core/model_registry.py,sha256=D0V95SBC0u5lJOJ0_kMqWJMtrH13ouxyCErQzp4Wdpo,15395
-isa_model/core/model_storage.py,sha256=yMLapW87EY1EPXw6S7H8UQAZh3hJ1KxsEohjgjw-HrA,4507
-isa_model/core/vision_models_init.py,sha256=wlv1hSs_KlN8wJCjtCDxJMeyaVpYsPDs3GSVw74mAJ0,4112
+isa_model/__init__.py,sha256=bAbHdrDVQ-mySC_GJtgfLNI8KPcs2LfReBkIOOtpaQQ,867
+isa_model/client.py,sha256=5u_hqGB1hcFX8MQdVYlCLqzs4ylQFY3rq91h3iTI24c,27500
+isa_model/core/config.py,sha256=h9GVTEEMlaJYSCDd0W9q1KtaWTV5V5TawMsKtGuphds,15686
+isa_model/core/pricing_manager.py,sha256=b7HcZsBQ8ZSCzMepOhqN-J9kU43vhTxX5NabQS0aM70,17125
+isa_model/core/types.py,sha256=XLUs442WGNc8E0gF2M-nb6dutD_s-XCfpr2BfGBCA2M,8445
+isa_model/core/config/__init__.py,sha256=SLeHQtYGDHl64NDVyb3ECQXOKepGM8YNHEoM8CVEWus,350
+isa_model/core/config/config_manager.py,sha256=hx0qcdvYEE_cCp-qb8tnVkXnpsTXEuRM1108DoAiUnE,19905
+isa_model/core/models/model_billing_tracker.py,sha256=hxqG5DvQ8kqROC3WVb5Y0tespp73TPPWYq-AwatMZwc,18782
+isa_model/core/models/model_manager.py,sha256=9_JHok4i2q79dhFD3nckrlG-k3z8UGb6bM82iUR-T84,14917
+isa_model/core/models/model_repo.py,sha256=OqvtUrvWmn_k4tkAgdOQ61NneUHsJGbUjkobJf3LNSw,14036
+isa_model/core/models/model_storage.py,sha256=yMLapW87EY1EPXw6S7H8UQAZh3hJ1KxsEohjgjw-HrA,4507
+isa_model/core/services/__init__.py,sha256=M1ob5N-PDPUW0o03ioHORRl5ye5ZEEzT7265-u-NrP4,481
+isa_model/core/services/intelligent_model_selector.py,sha256=vf8BhITetIh5VJcccsESrHyeh9p-V6AHD9NRAEt6a5w,21501
 isa_model/core/storage/hf_storage.py,sha256=HTj1-YGJM3Q-9_Adw7u4NjEmSdr0njsFEL45KXzfcFw,14701
 isa_model/core/storage/local_storage.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 isa_model/core/storage/minio_storage.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-isa_model/core/storage/supabase_storage.py,sha256=gW8EMhQGsodnPYqzDAKedLKOc7EZZ6uuuJ2B4fn20OU,14284
-isa_model/deployment/__init__.py,sha256=Wu-sBbQtwx7wzWu_MRON3RPmI4y8UfhK8pGe12-iUzs,1323
+isa_model/deployment/__init__.py,sha256=tzBOOvDXl6Hh8IiS2ehW6RKeUcBia_MY1uegQFFfFRg,1400
 isa_model/deployment/cloud/__init__.py,sha256=ztahTvIZYNoEU_bqmVAvb3xL_ttWHAGO_UAiwQP4dHY,135
 isa_model/deployment/cloud/modal/__init__.py,sha256=iD65eaqvI7L_W31p-tDO4fMIG5EWfkir1l3SKRSp72Y,265
-isa_model/deployment/cloud/modal/isa_vision_doc_service.py,sha256=5XHnX8BCILlzGEK53-WY7S1WJHSfiZv_qDJJw-9-4sc,23744
-isa_model/deployment/cloud/modal/isa_vision_ui_service.py,sha256=drCoK-ZZWg685FPnhY25X4OImsGYarjxIJAF3Mm8SnE,10318
+isa_model/deployment/cloud/modal/isa_vision_doc_service.py,sha256=GnIOwMNibXFqeA4ToUOovPxuj0NQTLXgkSbR1r6nUcU,30386
+isa_model/deployment/cloud/modal/isa_vision_table_service.py,sha256=ewuRLIM1sqeet_7Vb9y1xcB1vXg2OAHVZNDfeFA_KHM,19814
+isa_model/deployment/cloud/modal/isa_vision_ui_service.py,sha256=75iTaucfLYvWBqAvNdSXJT80R7hInYWOeLamxCyfteQ,14027
+isa_model/deployment/cloud/modal/register_models.py,sha256=gqvKDE8hdwOlSyg5nWrAYNlQ0AjHa4aP9OhFBw7UXLw,12162
 isa_model/deployment/core/__init__.py,sha256=QJkJrs0FYgYcjvnHMSvAtUBiT6uq_ciqLKWLwx0mkDg,803
 isa_model/deployment/core/deployment_config.py,sha256=__bHohsvbdZK_rS_86S1rSHPPP1bTkOnx_G0cj1HMcA,11305
 isa_model/deployment/core/deployment_manager.py,sha256=kICHX1V8wIlmldkrfdqakz2OAjitUfGY6ZG_QjGzZbM,20068
@@ -21,14 +28,26 @@ isa_model/deployment/core/isa_deployment_service.py,sha256=sXU7REZ4xhUUGrpqxlJh-
 isa_model/deployment/gpu_int8_ds8/app/server.py,sha256=lwWxdnR2DNEd0vIGQyfabKtDSUzSHVQsy3Z_AJejpVg,2102
 isa_model/deployment/gpu_int8_ds8/scripts/test_client.py,sha256=aCULgRYzEQj_ELUK1bmPgN99yvFgNR5C0O3gc8S32pg,1421
 isa_model/deployment/gpu_int8_ds8/scripts/test_client_os.py,sha256=XXrneTCHUeh1LNRcu-YtZQ5B4pNawlrxC-cTWmJU2A8,936
-isa_model/eval/__init__.py,sha256=3sM7qLSIL_RMKcsmkCYcjOjv9ozuk16r7pnl4F-XeNA,1197
+isa_model/deployment/runtime/deployed_service.py,sha256=0Z_Hg42oXasEVvuKjwBylJPazcmJYXhS-L9uPainaIg,13400
+isa_model/deployment/services/__init__.py,sha256=JrLlmBlLb6RfiqGMzVVxKZfF5tAKliQqpon_rPoNoeA,216
+isa_model/deployment/services/auto_deploy_vision_service.py,sha256=Luo8FaXsEBoKjGw5HQ7veOnv9Eh0e7U0claXaGa3-1o,19624
+isa_model/deployment/services/model_service.py,sha256=_ncC--8hr5BUwzCWh59yRXPKIPVLapx_31TorB2DIr8,13492
+isa_model/deployment/services/service_monitor.py,sha256=P1zGoeqkNEJwt9AXZF2qTjfSLRm5PKUa80GJVNDSIdA,15223
+isa_model/deployment/services/service_registry.py,sha256=LQgWQOvoP0lb7mC6WTS6shEt6WuX6xc8rRmcixrKwTc,22765
+isa_model/eval/__init__.py,sha256=CRbxC5SN7ow4ymdALSNTHawqw4f82DEdAb7twNT_Pw0,2447
 isa_model/eval/benchmarks.py,sha256=_L4Vwj2hwf2yhqoleIASO9z5e3LRCClCVEVCQbGt0I8,16885
-isa_model/eval/factory.py,sha256=uQXD1cZGPaMss2YGwtr8xONK9i_K7kHZG7-uwvNgEpk,29416
-isa_model/eval/metrics.py,sha256=mYeGwSa9PkgY0p-vadAscvak-pLrVfCSrsmAodVpgNQ,22584
+isa_model/eval/factory.py,sha256=bm5OVY7HIxdBgjlH1n7e5K1YO4ytv8e4KB7z_JS9HVQ,20737
+isa_model/eval/metrics.py,sha256=VMQqcB13OScJsEmfuMZeBk5RLCs01zWknuMqQRVFCjo,30315
+isa_model/eval/config/__init__.py,sha256=SDwprIRp1GSG7PAXJuyNldUZ6dPeuIcwdxCkHZXIeVQ,181
+isa_model/eval/config/evaluation_config.py,sha256=y4tQAqafx5snoQ0QD239C9QXHmqcQvjH34uGSEJUhvI,3051
+isa_model/eval/evaluators/__init__.py,sha256=bDMBhD8bTW740dQEt8Ru-nWPGiJHkyQQcfGEejgzkh4,461
+isa_model/eval/evaluators/base_evaluator.py,sha256=1RivgnnuA-k9EtsZ-KMEtts_5gcfnaTGAoF-uBVSesE,19292
+isa_model/eval/evaluators/llm_evaluator.py,sha256=yfFJFdxwGV2F3mzEWjZ-0fr9u8SR3A20UJ7zS7OgKsw,18054
+isa_model/eval/infrastructure/__init__.py,sha256=fxTdtwAFtjCDOV9MJ3GbhY0A-DqKeTwr_u9WTBnGI_U,648
+isa_model/eval/infrastructure/experiment_tracker.py,sha256=yfMWIAk6oA8Lfer3AtmKg0OEZiGhczmsCD5gmp--uew,15283
 isa_model/inference/__init__.py,sha256=usfuQJ4zYY2RRtHkE-V6LuJ5aN7WJogtPUj9Qmy4Wvw,318
-isa_model/inference/ai_factory.py,sha256=iSMHLMJ4iiPVCgzldjTdRS-2F0PsWK7Qlnjn9xQFtKk,33939
+isa_model/inference/ai_factory.py,sha256=IN-q3jNmcraZ-PWNTHyhdABoxxhIweZYcXO2fr_uXoM,16478
 isa_model/inference/base.py,sha256=qwOddnSGI0GUdD6qIdGBPQpkW7UjU3Y-zaZvu70B4WA,1278
-isa_model/inference/billing_tracker.py,sha256=uimayifP3oBZfU03qgveArJGl-1u6Vw2VTPj40O27t8,14888
 isa_model/inference/adapter/unified_api.py,sha256=67_Ok8W20m6Otf6r9WyOEVpnxondP4UAxOASk9ozDk4,8668
 isa_model/inference/providers/__init__.py,sha256=a83q-LMFv8u47wf0XtxvqOw_mlVgA_90wtuwy02qdDE,581
 isa_model/inference/providers/base_provider.py,sha256=PT-YnGwBu-Jn_4T3iAphkAJw_mYmKVLjUID62vf2_Ow,2711
@@ -41,59 +60,64 @@ isa_model/inference/providers/replicate_provider.py,sha256=0oi_BglIE6-HYgzLau9if
 isa_model/inference/providers/triton_provider.py,sha256=GKlth7cTOx6ERbsXXJ0gDNby3kVGQNULBDt098BXBSU,15258
 isa_model/inference/providers/yyds_provider.py,sha256=KbDsopShs11_G9oX3b2i2NgHIqkZV7HYWe9K9uZLccc,4284
 isa_model/inference/services/__init__.py,sha256=yfLz0YGl8ixk6LfTRL6cRTvZMb9F_Pv1QRgGyNc9xYM,386
-isa_model/inference/services/base_service.py,sha256=eTPlUAGAokEqruW9sXYN4A7jnfxfSSVsk2J28VhQRQo,3469
-isa_model/inference/services/audio/base_stt_service.py,sha256=OP2kFU5ZZT8yMpcbD3dpuCVzYOryY9XjQqAdalaFeYc,3347
-isa_model/inference/services/audio/base_tts_service.py,sha256=BzZ3JrrLpm4COthNyNrIO2QgP7RZkXDNPEELEKHzIbA,4164
+isa_model/inference/services/base_service.py,sha256=fVaSx0CogHK71UEsNJeSyM8mhqmq5_9ePbbSZVi3Al8,5085
+isa_model/inference/services/audio/base_stt_service.py,sha256=sfzAfreFdvEOBHtphoTrQSjb-gCoCOW4WCj6iIe51oU,5804
+isa_model/inference/services/audio/base_tts_service.py,sha256=PgctcV98Pe9I2kSjScsm8epRwdaEU-vAGCIfdd2P8us,6924
 isa_model/inference/services/audio/openai_realtime_service.py,sha256=UENsx1bEb7aJoXNuBtFGIbTmETpNTZcCHlv0RydEp_U,13340
-isa_model/inference/services/audio/openai_stt_service.py,sha256=S1BEhEctj5Aw86edn1Nd38tqB-cWq2RdiGxWCzhH2a8,10225
-isa_model/inference/services/audio/openai_tts_service.py,sha256=KtzcvS_szB_Xo26Lu3T9-Wc641aiyppCfyX4xUsYhSw,8132
-isa_model/inference/services/audio/replicate_tts_service.py,sha256=bW074ohLM4QQp0eaDq395_2hggUVMVAHoVVbSf75bvI,10000
-isa_model/inference/services/embedding/base_embed_service.py,sha256=Nr6snNtOM0_ZqFfJdV7ThTb2nYVHYddGoOJXkGuyBIg,3259
-isa_model/inference/services/embedding/ollama_embed_service.py,sha256=s6LPSh-D06kFYXQjoKJp8jnatW5cx_unGbVFaq7tm5c,4745
-isa_model/inference/services/embedding/openai_embed_service.py,sha256=cjAzmYXlY0XLgwemdlhhUVlDecgzFy2xn3DSycoYvdo,8474
-isa_model/inference/services/llm/__init__.py,sha256=C6t9w33j3Ap4oGcJal9-htifKe0rxwws_kC3F-_B_Ps,341
-isa_model/inference/services/llm/base_llm_service.py,sha256=rPVNq6UoTcuO387UzCxXxK_0BFFbd_t4BKcQB6TYU5g,6300
-isa_model/inference/services/llm/llm_adapter.py,sha256=gkKND55EoizRxsaGOCEazUmL-2yTIBI-_njpEOXt-4k,21779
-isa_model/inference/services/llm/ollama_llm_service.py,sha256=NXEStAL-A5znG9Ig62uwd9tpGttjxb8fOAhgRueJvzY,14696
-isa_model/inference/services/llm/openai_llm_service.py,sha256=h-hg_NBEf2ADSjAXeiIlSLv9uuLikAyOUunG-Aw_9Hg,10167
-isa_model/inference/services/llm/triton_llm_service.py,sha256=ZFo7JoZ799Nvyi8Cz1jfWOa6TUn0hDRJtBrotadMAd4,17673
-isa_model/inference/services/llm/yyds_llm_service.py,sha256=tKje8H-RpP_uAIrgKbEtA5k_uhY1ELW1BMH352vaMGk,10280
+isa_model/inference/services/audio/openai_stt_service.py,sha256=chqVuSU8JtwkqY6Y7INm0AdSoDqh-bsg9POzJkxN3h0,10989
+isa_model/inference/services/audio/openai_tts_service.py,sha256=C4vIRvCKoySs4-zBEteI_DZYZsATS84W_ZUwbxjJjpA,8253
+isa_model/inference/services/audio/replicate_tts_service.py,sha256=kCG_bBNgW7GQwt5-ZdwPSqsMiTV54-FhSowFwNWGvg0,10292
+isa_model/inference/services/embedding/base_embed_service.py,sha256=_9HLzwDfKXbhFWT_3VbLQpDF3N1Rr3JS-QXqAZM9Wyc,6045
+isa_model/inference/services/embedding/ollama_embed_service.py,sha256=Pbc3jePaEsudHOn07BSf_A525weoesH3d3r5vs8ODvc,6702
+isa_model/inference/services/embedding/openai_embed_service.py,sha256=Egzq0sxFWf5lndf12xijLuvJSw1wnLeU9wdNR9pnzkQ,8342
+isa_model/inference/services/embedding/helpers/text_splitter.py,sha256=6AbvcQ7H6MS54B9d9T1XBGg4GhvmKfZqp00lKp9pF-U,1635
+isa_model/inference/services/helpers/stacked_config.py,sha256=xPGvWBu364qSF5RXAefAdsqC9SUHC6frTtJamQRQZX8,5473
+isa_model/inference/services/img/__init__.py,sha256=Mip969-_UGRAADfDlcXEBPrsYY7sZleztolcjjmKfyA,518
+isa_model/inference/services/img/base_image_gen_service.py,sha256=PECzUood4E25Bh2hAbe-d62nB6CwRbg1pORdDkLlEjg,7194
+isa_model/inference/services/img/flux_professional_service.py,sha256=c-3F9ANUuYiZPG4SgsOgb6jTDFIXWD52Kz-RC9g_a-Q,26407
+isa_model/inference/services/img/replicate_image_gen_service.py,sha256=QLjgrXNmtdXmKQPCLkmrTjNPL7_icy6enWmeArRCicU,17240
+isa_model/inference/services/img/helpers/base_stacked_service.py,sha256=Y2g2Hz3n2Uwo4RoCbsC33IXxTeQ2TeT6T-2HnXd-2B0,10546
+isa_model/inference/services/llm/__init__.py,sha256=hSdfeqE0463CtQ6ZzREqPBvtzXNuzi3W5PFKS0nwhvo,336
+isa_model/inference/services/llm/base_llm_service.py,sha256=fg79J5mKy27qahvtLscMcquOe4O2L-EoNjE33yPRI9E,22196
+isa_model/inference/services/llm/ollama_llm_service.py,sha256=c0i9TLce6YcPlsLpj11kvwd-H2Uvw23G0ACuNShGyAM,16009
+isa_model/inference/services/llm/openai_llm_service.py,sha256=yhqsExCWvVs6G7EhwZnoXMtKncpwY1VPeZqgLibMSX0,11922
+isa_model/inference/services/llm/yyds_llm_service.py,sha256=WA5vwK9j7PrEEffdvP4thNwBlZ5Z13vFXX_F3Kd5V-Y,10475
+isa_model/inference/services/llm/helpers/llm_adapter.py,sha256=gkKND55EoizRxsaGOCEazUmL-2yTIBI-_njpEOXt-4k,21779
+isa_model/inference/services/llm/helpers/llm_prompts.py,sha256=4ZOovr_Jx5bwPiMkgO1lZF7GMyEFbITP3ZrhhI48QMs,10964
+isa_model/inference/services/llm/helpers/llm_utils.py,sha256=LNfB3qzREyD2TyleOyVa20FU8HbhdxA8G6QV-N-wfZs,10016
 isa_model/inference/services/ml/base_ml_service.py,sha256=mLBA6ENowa3KVzNqHyhWxf_Pr-cJJj84lDE4TniPzYI,2894
 isa_model/inference/services/ml/sklearn_ml_service.py,sha256=Lf9JrwvI25lca7JBbjB_e66eAUtXFbwxZ3Hs13dVGkA,5512
 isa_model/inference/services/others/table_transformer_service.py,sha256=r74h6QUSwSj6jTt-gRProz9SgwBwKWDe50NR0uqW0ZI,2367
-isa_model/inference/services/stacked/__init__.py,sha256=ud8OgxsykEUzB3hGV-2heEYh1kXVOgiF9IYj5UvdSHs,824
-isa_model/inference/services/stacked/base_stacked_service.py,sha256=13Pb2FiB3iLHft0l27ril8czY6Pznti8UY-frN5gCB0,10546
-isa_model/inference/services/stacked/config.py,sha256=BCJhZQNwC4vP2FA6eKogRJ1TW6O_whuYkqDdmciWqac,16758
-isa_model/inference/services/stacked/doc_analysis_service.py,sha256=q6yB_RU3C2l94S49XKPOZjd0LAnRRpYhckRuFuyn8fs,24825
-isa_model/inference/services/stacked/flux_professional_service.py,sha256=DwfcSNSt3sahbzaDFzci2hsnUoAqKbZReL1fvwchzU4,25297
-isa_model/inference/services/stacked/ui_analysis_service.py,sha256=5-8qS6hZSRkEFluVrvU0K7EvH0IWUVLxMQBjk0Juolg,52266
-isa_model/inference/services/vision/__init__.py,sha256=N9Zr7o2uQKoyUEvpmyOIgXPx9ivrix3gQ1OLoiQ7BLo,283
-isa_model/inference/services/vision/base_image_gen_service.py,sha256=k88mWy5yReqMbz-ZT8aK9P3Bbh2KaykJV9p_MB6DD2g,4247
-isa_model/inference/services/vision/base_vision_service.py,sha256=ismIKNxZa0TQi0KSyEXewubjgMERl_5Pzqu4eQih7Qg,6840
-isa_model/inference/services/vision/isA_vision_service.py,sha256=9lgxW83CInwmndtI_9-4iRMhcl0vG-sSx3xLfNnfojs,14620
-isa_model/inference/services/vision/ollama_vision_service.py,sha256=Btm3jJmnSBcJDiTujr51eWC3a3eA_58xKMj5TsatXJQ,6821
-isa_model/inference/services/vision/openai_vision_service.py,sha256=aGjYI5jRzL73hchFFLb1077rnnoUFDJ_EEV14EjqJG8,18686
-isa_model/inference/services/vision/replicate_image_gen_service.py,sha256=NMOPNTLKOJi_SF86pI5mgvGIO1IMEEiU9TmBbRcfLXM,16509
-isa_model/inference/services/vision/replicate_vision_service.py,sha256=UtuC_jrSD_ZYwLbrRZPzTo5ElXHQNYyAywcwTHaXpTM,28474
-isa_model/inference/services/vision/helpers/image_utils.py,sha256=ieEL69LQ9-T4zsSFj2Mmt2jRUU_UOUAgt1W6Je9kaa8,1800
-isa_model/inference/services/vision/helpers/text_splitter.py,sha256=6AbvcQ7H6MS54B9d9T1XBGg4GhvmKfZqp00lKp9pF-U,1635
+isa_model/inference/services/vision/__init__.py,sha256=9JkNzwZa6WQPUXtq4i7x78mPxkzBNRB-6CkU2WyoCYE,1191
+isa_model/inference/services/vision/base_vision_service.py,sha256=FInmrdhKaeu0vBUbxScH3_B9a_fk3pj2foLx0P81a6o,12645
+isa_model/inference/services/vision/doc_analysis_service.py,sha256=_klnuI99tSkA1gcs5hPQmj6Ld8GElg0S823LXa9id2Y,24797
+isa_model/inference/services/vision/openai_vision_service.py,sha256=IcDX0nBPqt9TJSwMEUD89Fojz1X8jF3ZD1voJGOzu4A,9609
+isa_model/inference/services/vision/replicate_vision_service.py,sha256=smRkSCTwk5mvyKVnvyplqPNuVYjRZngVBWxTCbFmrxA,20679
+isa_model/inference/services/vision/ui_analysis_service.py,sha256=LTFY5rMIQwgWh_kh9KWA44ZI01Mb3We5NsG2scSgBeA,32153
+isa_model/inference/services/vision/disabled/isA_vision_service.py,sha256=VYa8VJtxDB9KdnfNW0GPEP_TPker4pHp33gLD_TnpaM,18336
+isa_model/inference/services/vision/helpers/base_stacked_service.py,sha256=Y2g2Hz3n2Uwo4RoCbsC33IXxTeQ2TeT6T-2HnXd-2B0,10546
+isa_model/inference/services/vision/helpers/image_utils.py,sha256=7RZ368o2UXcRCqyzldUOGhzB5vWbCCJOVR24UdiVdhs,11061
+isa_model/inference/services/vision/helpers/vision_prompts.py,sha256=WbzOYu-Z2-8Xn9dcvuPRTA7VTy23_uoMRRGO4t0wZ8Q,12098
 isa_model/inference/utils/conversion/bge_rerank_convert.py,sha256=1dvtxe5-PPCe2Au6SO8F2XaD-xdIoeA4zDTcid2L9FU,2691
 isa_model/inference/utils/conversion/onnx_converter.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 isa_model/inference/utils/conversion/torch_converter.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 isa_model/scripts/inference_tracker.py,sha256=T6qQJMHJcAIQ8eYlgqpM9RWxfiV4z5xIolaoglKBBsg,8831
 isa_model/scripts/mlflow_manager.py,sha256=7xMN0_wELr1jcALuTW9WeWirRkPZPlE2LlFfZKflXBY,12142
 isa_model/scripts/model_registry.py,sha256=7rycPkVk8WHUO3LJaHfdyy5Yq8qmd_4WkGk4wKan-2w,14279
+isa_model/scripts/register_models.py,sha256=ukqhUFiXkZh2IGzJTjEKpqRpf6VW4oSYm4PP_JjYcM8,14658
+isa_model/scripts/register_models_with_embeddings.py,sha256=A4utiT5_uqvgMQJ9qioIZM2XMreHLgJoqZOysNbzmM8,20990
 isa_model/scripts/start_mlflow.py,sha256=3AGKBzByjzbZ56I8w0IOfYnp3V6EU2Lv9NtX9maSqL8,2571
 isa_model/scripts/training_tracker.py,sha256=cnXPi8ip2OK76-aWAOgC-dKx90PqZLEnP6UbHso7Fwc,8080
 isa_model/serving/__init__.py,sha256=LTO0Adbvm7A-bgQqtuOQSoHvdu9OH3OrEjYgQanuHgI,429
 isa_model/serving/api/__init__.py,sha256=wgWD69eqV37fFTLxhz8b0rOn_34P7SZHoWw2sufWjk4,162
-isa_model/serving/api/fastapi_server.py,sha256=Q9rvdyAyDD7fMqZ-EusOWYXPXyvce86a7_wKuTwLvFg,2403
+isa_model/serving/api/fastapi_server.py,sha256=UP3CZim8E2tNcyUGaoU9cYM80_yzzpWQFd3ADfo6ANk,2628
 isa_model/serving/api/middleware/__init__.py,sha256=iCKUYECf0bjNGXgV91K03hb8Dnp0Jc_wnUL897Rd0sg,163
 isa_model/serving/api/middleware/request_logger.py,sha256=d48n6tp1pqZ7HFWFl8jg6er24ugWkWkMOc1y80aqPU8,2938
 isa_model/serving/api/routes/__init__.py,sha256=RIaG9OPg0AjAIVbtMzwnqGyNU-tuQXbdvFcYOt4b_Do,84
 isa_model/serving/api/routes/health.py,sha256=NwQcC_bpcaI4YZHTIKbGtg82yQ6QLdp0TwcqbEiqbWs,2208
 isa_model/serving/api/routes/llm.py,sha256=5ZVxWugff0i6VBKz_Nv5CqacMZJsPZEKyoSB6XDrW34,385
 isa_model/serving/api/routes/ui_analysis.py,sha256=-WxLaRKQNHnRh4okB85cWA4blTegpEPZtzHTsF3yeeU,6848
+isa_model/serving/api/routes/unified.py,sha256=rSoHPtMWkGlzFwzzoZeFlCmFGWj2r3q-5QH9VeNQQxA,7074
 isa_model/serving/api/routes/vision.py,sha256=U9jxssQYe6igtayUW0C2fcYwqmLRIE15__X-5Ru9J4c,396
 isa_model/serving/api/schemas/__init__.py,sha256=Tu_hzxoKW1ZHpww3-5ER4A2hNuDByZ0rAfrgaJ7Bs-M,275
 isa_model/serving/api/schemas/common.py,sha256=HVaAS7wlvqrwC1gMZ2Cvo0vzHB053x2uOTAwUoY2vsE,696
@@ -117,7 +141,7 @@ isa_model/training/core/config.py,sha256=oqgKpBvtzrN6jwLIQYQ2707lH6nmjrktRiSxp9i
 isa_model/training/core/dataset.py,sha256=XCFsnf0NUMU1dJpdvo_CAMyvXB-9_RCUEiy8TU50e20,7802
 isa_model/training/core/trainer.py,sha256=h5TjqjdFr0Fsv5y4-0siy1KmOlqLfliVaUXybvuoeXU,26932
 isa_model/training/core/utils.py,sha256=Nik0M2ssfNbWqP6fKO0Kfyhzr_H6Q19ioxB-qCYbn5E,8387
-isa_model-0.3.5.dist-info/METADATA,sha256=jhARc6WBDaFWWScGhArHukMp4m1J4iJlo7KlGMIl2p0,12226
-isa_model-0.3.5.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-isa_model-0.3.5.dist-info/top_level.txt,sha256=eHSy_Xb3kNkh2kK11mi1mZh0Wz91AQ5b8k2KFYO-rE8,10
-isa_model-0.3.5.dist-info/RECORD,,
+isa_model-0.3.6.dist-info/METADATA,sha256=TMGcK76gGTCDWcXfCp17JuAoWxVN4TfVv5Nu-8mN8JE,12326
+isa_model-0.3.6.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+isa_model-0.3.6.dist-info/top_level.txt,sha256=eHSy_Xb3kNkh2kK11mi1mZh0Wz91AQ5b8k2KFYO-rE8,10
+isa_model-0.3.6.dist-info/RECORD,,

isa_model/config/__init__.py

@@ -1,9 +0,0 @@
-"""
-Configuration Management Module
-
-Unified configuration system for all ISA Model components
-"""
-
-from .config_manager import ConfigManager, get_config, DeploymentConfig, ModelConfig
-
-__all__ = ["ConfigManager", "get_config", "DeploymentConfig", "ModelConfig"]

isa_model/config/config_manager.py

@@ -1,213 +0,0 @@
-"""
-Configuration Manager
-
-Central configuration management with environment support
-"""
-
-import os
-import yaml
-from typing import Dict, Any, Optional
-from pathlib import Path
-from dataclasses import dataclass
-import logging
-
-logger = logging.getLogger(__name__)
-
-@dataclass
-class ConfigSection:
-    """Base configuration section"""
-    def to_dict(self) -> Dict[str, Any]:
-        return self.__dict__
-
-@dataclass 
-class DeploymentConfig(ConfigSection):
-    """Deployment configuration"""
-    platform: str = "replicate"  # replicate, modal, aws, local
-    modal_app_name: str = "isa-ui-analysis"
-    modal_gpu_type: str = "A100-40GB"
-    modal_memory: int = 32768
-    modal_timeout: int = 1800
-    modal_keep_warm: int = 1
-    
-@dataclass
-class ModelConfig(ConfigSection):
-    """Model configuration"""
-    ui_detection_model: str = "microsoft/omniparser-v2"
-    ui_planning_model: str = "gpt-4o-mini"
-    fallback_detection: str = "yolov8n"
-    quantization: bool = False
-    batch_size: int = 1
-
-@dataclass
-class ServingConfig(ConfigSection):
-    """Serving configuration"""
-    host: str = "0.0.0.0"
-    port: int = 8000
-    workers: int = 1
-    reload: bool = False
-    log_level: str = "info"
-    cors_origins: list = None
-    
-    def __post_init__(self):
-        if self.cors_origins is None:
-            self.cors_origins = ["*"]
-
-@dataclass
-class APIConfig(ConfigSection):
-    """API configuration"""
-    rate_limit: int = 100  # requests per minute
-    max_file_size: int = 10 * 1024 * 1024  # 10MB
-    cache_ttl: int = 3600  # 1 hour
-    enable_auth: bool = False
-
-@dataclass
-class ISAConfig:
-    """Complete ISA configuration"""
-    environment: str
-    deployment: DeploymentConfig
-    models: ModelConfig  
-    serving: ServingConfig
-    api: APIConfig
-    
-    def to_dict(self) -> Dict[str, Any]:
-        return {
-            "environment": self.environment,
-            "deployment": self.deployment.to_dict(),
-            "models": self.models.to_dict(),
-            "serving": self.serving.to_dict(),
-            "api": self.api.to_dict()
-        }
-
-class ConfigManager:
-    """Configuration manager with environment support"""
-    
-    _instance = None
-    _config = None
-    
-    def __new__(cls):
-        if cls._instance is None:
-            cls._instance = super(ConfigManager, cls).__new__(cls)
-        return cls._instance
-    
-    def __init__(self):
-        if self._config is None:
-            self._load_config()
-    
-    def _load_config(self):
-        """Load configuration from environment and files"""
-        env = os.getenv("ISA_ENV", "development")
-        
-        # Default configurations
-        default_config = {
-            "deployment": DeploymentConfig(),
-            "models": ModelConfig(), 
-            "serving": ServingConfig(),
-            "api": APIConfig()
-        }
-        
-        # Load environment-specific configuration
-        config_file = self._get_config_file(env)
-        if config_file and config_file.exists():
-            try:
-                with open(config_file, 'r') as f:
-                    file_config = yaml.safe_load(f)
-                    
-                # Merge configurations
-                self._config = self._merge_configs(default_config, file_config, env)
-                logger.info(f"Loaded configuration for environment: {env}")
-                
-            except Exception as e:
-                logger.warning(f"Failed to load config file {config_file}: {e}")
-                self._config = ISAConfig(environment=env, **default_config)
-        else:
-            logger.info(f"No config file found for {env}, using defaults")
-            self._config = ISAConfig(environment=env, **default_config)
-        
-        # Override with environment variables
-        self._apply_env_overrides()
-    
-    def _get_config_file(self, env: str) -> Optional[Path]:
-        """Get configuration file path for environment"""
-        # Try to find config file in multiple locations
-        possible_paths = [
-            Path(__file__).parent / "environments" / f"{env}.yaml",
-            Path.cwd() / "config" / f"{env}.yaml",
-            Path.cwd() / f"config_{env}.yaml"
-        ]
-        
-        for path in possible_paths:
-            if path.exists():
-                return path
-        return None
-    
-    def _merge_configs(self, default: Dict, file_config: Dict, env: str) -> ISAConfig:
-        """Merge default and file configurations"""
-        
-        # Update deployment config
-        deployment_data = {**default["deployment"].__dict__}
-        if "deployment" in file_config:
-            deployment_data.update(file_config["deployment"])
-        deployment = DeploymentConfig(**deployment_data)
-        
-        # Update model config
-        models_data = {**default["models"].__dict__}
-        if "models" in file_config:
-            models_data.update(file_config["models"])
-        models = ModelConfig(**models_data)
-        
-        # Update serving config
-        serving_data = {**default["serving"].__dict__}
-        if "serving" in file_config:
-            serving_data.update(file_config["serving"])
-        serving = ServingConfig(**serving_data)
-        
-        # Update API config
-        api_data = {**default["api"].__dict__}
-        if "api" in file_config:
-            api_data.update(file_config["api"])
-        api = APIConfig(**api_data)
-        
-        return ISAConfig(
-            environment=env,
-            deployment=deployment,
-            models=models,
-            serving=serving,
-            api=api
-        )
-    
-    def _apply_env_overrides(self):
-        """Apply environment variable overrides"""
-        # Deployment overrides
-        if os.getenv("ISA_DEPLOYMENT_PLATFORM"):
-            self._config.deployment.platform = os.getenv("ISA_DEPLOYMENT_PLATFORM")
-        
-        # Model overrides
-        if os.getenv("ISA_UI_DETECTION_MODEL"):
-            self._config.models.ui_detection_model = os.getenv("ISA_UI_DETECTION_MODEL")
-        
-        # Serving overrides
-        if os.getenv("ISA_SERVING_PORT"):
-            self._config.serving.port = int(os.getenv("ISA_SERVING_PORT"))
-        
-        if os.getenv("ISA_SERVING_HOST"):
-            self._config.serving.host = os.getenv("ISA_SERVING_HOST")
-    
-    def get_config(self) -> ISAConfig:
-        """Get current configuration"""
-        return self._config
-    
-    def reload(self):
-        """Reload configuration"""
-        self._config = None
-        self._load_config()
-
-# Singleton instance
-_config_manager = ConfigManager()
-
-def get_config() -> ISAConfig:
-    """Get configuration instance"""
-    return _config_manager.get_config()
-
-def reload_config():
-    """Reload configuration"""
-    _config_manager.reload()