isa-model 0.4.3__tar.gz → 0.4.4__tar.gz

{isa_model-0.4.3 → isa_model-0.4.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: isa_model
-Version: 0.4.3
+Version: 0.4.4
 Summary: Unified AI model serving framework
 Author: isA_Model Contributors
 Classifier: Development Status :: 3 - Alpha
@@ -89,6 +89,11 @@ Requires-Dist: isa-model[cloud,k8s,monitoring,storage]; extra == "production"
 Provides-Extra: staging
 Requires-Dist: isa-model[cloud,langchain,monitoring,storage]; extra == "staging"
 Requires-Dist: python-consul>=1.1.0; extra == "staging"
+Provides-Extra: staging-minimal
+Requires-Dist: isa-model[cloud,langchain,storage]; extra == "staging-minimal"
+Requires-Dist: influxdb-client>=1.36.0; extra == "staging-minimal"
+Requires-Dist: python-logging-loki>=0.3.1; extra == "staging-minimal"
+Requires-Dist: python-consul>=1.1.0; extra == "staging-minimal"
 Provides-Extra: all
 Requires-Dist: isa-model[audio,cloud,gpu-cloud,k8s,langchain,local,monitoring,storage,training,vision]; extra == "all"
 

{isa_model-0.4.3 → isa_model-0.4.4}/isa_model/core/config.py

@@ -54,7 +54,7 @@ class LocalGPUGlobalConfig:
     enable_local_gpu: bool = True
     auto_detect_gpu: bool = True
     workspace_dir: str = "./local_deployments"
-    preferred_backend: str = "transformers"  # vllm, tensorrt_llm, transformers
+    preferred_backend: str = "api"  # cloud api only
     
     # Default resource settings
     default_gpu_memory_fraction: float = 0.9
@@ -114,7 +114,7 @@ class GlobalConfig:
     local_gpu_memory_fraction: float = 0.9
     local_workspace_dir: str = "./local_deployments"
     auto_detect_gpu: bool = True
-    preferred_local_backend: str = "transformers"  # vllm, tensorrt_llm, transformers
+    preferred_local_backend: str = "api"  # cloud api only
     
     # Local service defaults
     local_health_check_interval: int = 30  # seconds
@@ -263,7 +263,7 @@ class ConfigManager:
             "enable_local_gpu": os.getenv("ISA_ENABLE_LOCAL_GPU", "true").lower() == "true",
             "auto_detect_gpu": os.getenv("ISA_AUTO_DETECT_GPU", "true").lower() == "true",
             "workspace_dir": os.getenv("ISA_LOCAL_WORKSPACE_DIR", "./local_deployments"),
-            "preferred_backend": os.getenv("ISA_PREFERRED_LOCAL_BACKEND", "transformers"),
+            "preferred_backend": os.getenv("ISA_PREFERRED_LOCAL_BACKEND", "api"),
             "default_gpu_memory_fraction": float(os.getenv("ISA_GPU_MEMORY_FRACTION", "0.9")),
             "health_check_interval": int(os.getenv("ISA_LOCAL_HEALTH_CHECK_INTERVAL", "30")),
             "max_concurrent_services": int(os.getenv("ISA_MAX_CONCURRENT_SERVICES", "3")),

{isa_model-0.4.3 → isa_model-0.4.4}/isa_model/core/models/model_manager.py

@@ -2,8 +2,6 @@ from typing import Dict, Optional, List, Any
 import logging
 from pathlib import Path
 from datetime import datetime
-from huggingface_hub import hf_hub_download, snapshot_download
-from huggingface_hub.errors import HfHubHTTPError
 from .model_storage import ModelStorage, LocalModelStorage
 from .model_repo import ModelRegistry, ModelType, ModelCapability
 from .model_billing_tracker import ModelBillingTracker, ModelOperationType
@@ -102,73 +100,7 @@ class ModelManager:
             logger.warning(f"Failed to find cheapest model for {provider}: {e}")
             return None
 
-    async def get_model(self, 
-                       model_id: str, 
-                       repo_id: str,
-                       model_type: ModelType,
-                       capabilities: List[ModelCapability],
-                       revision: Optional[str] = None,
-                       force_download: bool = False) -> Optional[Path]:
-        """
-        Get model files, downloading if necessary
-        
-        Args:
-            model_id: Unique identifier for the model
-            repo_id: Hugging Face repository ID
-            model_type: Type of model (LLM, embedding, etc.)
-            capabilities: List of model capabilities
-            revision: Specific model version/tag
-            force_download: Force re-download even if cached
-            
-        Returns:
-            Path to the model files or None if failed
-        """
-        # Check if model is already downloaded
-        if not force_download:
-            model_path = await self.storage.load_model(model_id)
-            if model_path:
-                logger.info(f"Using cached model {model_id}")
-                return model_path
-        
-        try:
-            # Download model files
-            logger.info(f"Downloading model {model_id} from {repo_id}")
-            model_dir = Path(f"./models/temp/{model_id}")
-            model_dir.mkdir(parents=True, exist_ok=True)
-
-            snapshot_download(
-                repo_id=repo_id,
-                revision=revision,
-                local_dir=model_dir,
-                local_dir_use_symlinks=False
-            )
-
-            # Save model and metadata
-            metadata = {
-                "repo_id": repo_id,
-                "revision": revision,
-                "downloaded_at": str(Path(model_dir).stat().st_mtime)
-            }
-
-            # Register model
-            self.registry.register_model(
-                model_id=model_id,
-                model_type=model_type,
-                capabilities=capabilities,
-                metadata=metadata
-            )
-
-            # Save model files
-            await self.storage.save_model(model_id, str(model_dir), metadata)
-
-            return await self.storage.load_model(model_id)
-
-        except HfHubHTTPError as e:
-            logger.error(f"Failed to download model {model_id}: {e}")
-            return None
-        except Exception as e:
-            logger.error(f"Unexpected error downloading model {model_id}: {e}")
-            return None
+    # Local model download functionality removed - use cloud API services only
     
     async def list_models(self) -> List[Dict[str, Any]]:
         """List all downloaded models with their metadata"""

{isa_model-0.4.3 → isa_model-0.4.4}/isa_model.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: isa_model
-Version: 0.4.3
+Version: 0.4.4
 Summary: Unified AI model serving framework
 Author: isA_Model Contributors
 Classifier: Development Status :: 3 - Alpha
@@ -89,6 +89,11 @@ Requires-Dist: isa-model[cloud,k8s,monitoring,storage]; extra == "production"
 Provides-Extra: staging
 Requires-Dist: isa-model[cloud,langchain,monitoring,storage]; extra == "staging"
 Requires-Dist: python-consul>=1.1.0; extra == "staging"
+Provides-Extra: staging-minimal
+Requires-Dist: isa-model[cloud,langchain,storage]; extra == "staging-minimal"
+Requires-Dist: influxdb-client>=1.36.0; extra == "staging-minimal"
+Requires-Dist: python-logging-loki>=0.3.1; extra == "staging-minimal"
+Requires-Dist: python-consul>=1.1.0; extra == "staging-minimal"
 Provides-Extra: all
 Requires-Dist: isa-model[audio,cloud,gpu-cloud,k8s,langchain,local,monitoring,storage,training,vision]; extra == "all"
 

{isa_model-0.4.3 → isa_model-0.4.4}/isa_model.egg-info/SOURCES.txt

@@ -40,22 +40,12 @@ isa_model/core/models/system_models.py
 isa_model/core/repositories/__init__.py
 isa_model/core/repositories/config_repository.py
 isa_model/core/resilience/circuit_breaker.py
-isa_model/core/security/secrets.py
 isa_model/core/services/__init__.py
 isa_model/core/services/intelligent_model_selector.py
-isa_model/core/storage/hf_storage.py
 isa_model/core/storage/local_storage.py
 isa_model/deployment/__init__.py
 isa_model/deployment/core/__init__.py
 isa_model/deployment/core/deployment_manager.py
-isa_model/deployment/local/__init__.py
-isa_model/deployment/local/config.py
-isa_model/deployment/local/gpu_gateway.py
-isa_model/deployment/local/health_checker.py
-isa_model/deployment/local/provider.py
-isa_model/deployment/local/tensorrt_service.py
-isa_model/deployment/local/transformers_service.py
-isa_model/deployment/local/vllm_service.py
 isa_model/deployment/modal/__init__.py
 isa_model/deployment/modal/config.py
 isa_model/deployment/modal/deployer.py
@@ -107,7 +97,6 @@ isa_model/inference/repositories/__init__.py
 isa_model/inference/repositories/inference_repository.py
 isa_model/inference/services/__init__.py
 isa_model/inference/services/base_service.py
-isa_model/inference/services/custom_model_manager.py
 isa_model/inference/services/audio/__init__.py
 isa_model/inference/services/audio/base_realtime_service.py
 isa_model/inference/services/audio/base_stt_service.py
@@ -137,7 +126,6 @@ isa_model/inference/services/llm/__init__.py
 isa_model/inference/services/llm/base_llm_service.py
 isa_model/inference/services/llm/cerebras_llm_service.py
 isa_model/inference/services/llm/huggingface_llm_service.py
-isa_model/inference/services/llm/local_llm_service.py
 isa_model/inference/services/llm/ollama_llm_service.py
 isa_model/inference/services/llm/openai_llm_service.py
 isa_model/inference/services/llm/yyds_llm_service.py
@@ -148,7 +136,6 @@ isa_model/inference/services/ml/base_ml_service.py
 isa_model/inference/services/ml/sklearn_ml_service.py
 isa_model/inference/services/vision/__init__.py
 isa_model/inference/services/vision/base_vision_service.py
-isa_model/inference/services/vision/blip_vision_service.py
 isa_model/inference/services/vision/isa_vision_service.py
 isa_model/inference/services/vision/openai_vision_service.py
 isa_model/inference/services/vision/replicate_vision_service.py

{isa_model-0.4.3 → isa_model-0.4.4}/isa_model.egg-info/requires.txt

@@ -83,6 +83,12 @@ isa-model[cloud,k8s,monitoring,storage]
 isa-model[cloud,langchain,monitoring,storage]
 python-consul>=1.1.0
 
+[staging-minimal]
+isa-model[cloud,langchain,storage]
+influxdb-client>=1.36.0
+python-logging-loki>=0.3.1
+python-consul>=1.1.0
+
 [storage]
 boto3>=1.26.0
 google-cloud-storage>=2.7.0

{isa_model-0.4.3 → isa_model-0.4.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "isa_model"
-version = "0.4.3"
+version = "0.4.4"
 description = "Unified AI model serving framework"
 authors = [{name = "isA_Model Contributors"}]
 readme = "README.md"
@@ -152,6 +152,14 @@ staging = [
     "python-consul>=1.1.0",  # Service discovery
 ]
 
+# Ultra-lightweight staging (no MLflow monitoring)
+staging-minimal = [
+    "isa-model[cloud,storage,langchain]",
+    "influxdb-client>=1.36.0",  # Basic logging only
+    "python-logging-loki>=0.3.1",  # Centralized logging
+    "python-consul>=1.1.0",  # Service discovery
+]
+
 all = [
     "isa-model[cloud,local,training,audio,vision,langchain,storage,monitoring,k8s,gpu-cloud]",
 ]

isa_model-0.4.3/isa_model/core/security/secrets.py

@@ -1,358 +0,0 @@
-"""
-Secrets Management System
-
-Provides secure handling of API keys, tokens, and other sensitive data.
-Supports multiple backends: environment variables, HashiCorp Vault, AWS Secrets Manager.
-"""
-
-import os
-import json
-import logging
-import hashlib
-import base64
-from typing import Dict, Optional, Any, List
-from pathlib import Path
-from cryptography.fernet import Fernet
-from cryptography.hazmat.primitives import hashes
-from cryptography.hazmat.primitives.kdf.pbkdf2 import PBKDF2HMAC
-import structlog
-
-logger = structlog.get_logger(__name__)
-
-class SecretsManager:
-    """Unified secrets management with multiple backend support"""
-    
-    def __init__(self, backend: str = "env", **kwargs):
-        self.backend = backend
-        self.config = kwargs
-        self._cache = {}
-        self._encryption_key = None
-        
-        # Initialize encryption key for local storage
-        self._init_encryption()
-        
-        # Initialize backend
-        if backend == "vault":
-            self._init_vault()
-        elif backend == "aws":
-            self._init_aws()
-        elif backend == "env":
-            self._init_env()
-        else:
-            raise ValueError(f"Unsupported secrets backend: {backend}")
-        
-        logger.info("Secrets manager initialized", backend=backend)
-    
-    def _init_encryption(self):
-        """Initialize encryption for local secret storage"""
-        # Use a combination of environment and system info for key derivation
-        password = os.getenv("SECRET_ENCRYPTION_KEY", "default-key-change-in-production").encode()
-        salt = os.getenv("SECRET_SALT", "default-salt").encode()
-        
-        kdf = PBKDF2HMAC(
-            algorithm=hashes.SHA256(),
-            length=32,
-            salt=salt,
-            iterations=100000,
-        )
-        key = base64.urlsafe_b64encode(kdf.derive(password))
-        self._encryption_key = Fernet(key)
-    
-    def _init_env(self):
-        """Initialize environment variable backend"""
-        logger.info("Using environment variables for secrets")
-    
-    def _init_vault(self):
-        """Initialize HashiCorp Vault backend"""
-        try:
-            import hvac
-            
-            vault_url = self.config.get("vault_url", os.getenv("VAULT_URL"))
-            vault_token = self.config.get("vault_token", os.getenv("VAULT_TOKEN"))
-            
-            if not vault_url:
-                raise ValueError("VAULT_URL required for Vault backend")
-            
-            self.vault_client = hvac.Client(url=vault_url, token=vault_token)
-            
-            if not self.vault_client.is_authenticated():
-                raise ValueError("Vault authentication failed")
-            
-            logger.info("Vault backend initialized", url=vault_url)
-            
-        except ImportError:
-            raise ImportError("hvac package required for Vault backend: pip install hvac")
-    
-    def _init_aws(self):
-        """Initialize AWS Secrets Manager backend"""
-        try:
-            import boto3
-            
-            region = self.config.get("region", os.getenv("AWS_REGION", "us-east-1"))
-            self.secrets_client = boto3.client("secretsmanager", region_name=region)
-            
-            logger.info("AWS Secrets Manager backend initialized", region=region)
-            
-        except ImportError:
-            raise ImportError("boto3 package required for AWS backend: pip install boto3")
-    
-    def get_secret(self, key: str, default: Optional[str] = None) -> Optional[str]:
-        """Get a secret value by key"""
-        # Check cache first
-        if key in self._cache:
-            return self._cache[key]
-        
-        try:
-            if self.backend == "env":
-                value = self._get_env_secret(key, default)
-            elif self.backend == "vault":
-                value = self._get_vault_secret(key, default)
-            elif self.backend == "aws":
-                value = self._get_aws_secret(key, default)
-            else:
-                value = default
-            
-            # Cache the value
-            if value is not None:
-                self._cache[key] = value
-            
-            return value
-            
-        except Exception as e:
-            logger.error("Failed to retrieve secret", key=key, error=str(e))
-            return default
-    
-    def _get_env_secret(self, key: str, default: Optional[str] = None) -> Optional[str]:
-        """Get secret from environment variables"""
-        return os.getenv(key, default)
-    
-    def _get_vault_secret(self, key: str, default: Optional[str] = None) -> Optional[str]:
-        """Get secret from HashiCorp Vault"""
-        try:
-            secret_path = self.config.get("secret_path", "secret/data/isa-model")
-            response = self.vault_client.secrets.kv.v2.read_secret_version(path=secret_path)
-            data = response["data"]["data"]
-            return data.get(key, default)
-        except Exception as e:
-            logger.warning("Failed to retrieve secret from Vault", key=key, error=str(e))
-            return default
-    
-    def _get_aws_secret(self, key: str, default: Optional[str] = None) -> Optional[str]:
-        """Get secret from AWS Secrets Manager"""
-        try:
-            secret_name = self.config.get("secret_name", "isa-model/secrets")
-            response = self.secrets_client.get_secret_value(SecretId=secret_name)
-            secrets = json.loads(response["SecretString"])
-            return secrets.get(key, default)
-        except Exception as e:
-            logger.warning("Failed to retrieve secret from AWS", key=key, error=str(e))
-            return default
-    
-    def set_secret(self, key: str, value: str) -> bool:
-        """Set a secret value (only supported for some backends)"""
-        try:
-            if self.backend == "vault":
-                return self._set_vault_secret(key, value)
-            elif self.backend == "aws":
-                return self._set_aws_secret(key, value)
-            else:
-                logger.warning("Set operation not supported for backend", backend=self.backend)
-                return False
-        except Exception as e:
-            logger.error("Failed to set secret", key=key, error=str(e))
-            return False
-    
-    def _set_vault_secret(self, key: str, value: str) -> bool:
-        """Set secret in HashiCorp Vault"""
-        try:
-            secret_path = self.config.get("secret_path", "secret/data/isa-model")
-            # Get existing secrets first
-            try:
-                response = self.vault_client.secrets.kv.v2.read_secret_version(path=secret_path)
-                existing_data = response["data"]["data"]
-            except:
-                existing_data = {}
-            
-            # Update with new secret
-            existing_data[key] = value
-            
-            # Write back to vault
-            self.vault_client.secrets.kv.v2.create_or_update_secret(
-                path=secret_path,
-                secret=existing_data
-            )
-            
-            # Update cache
-            self._cache[key] = value
-            return True
-            
-        except Exception as e:
-            logger.error("Failed to set secret in Vault", key=key, error=str(e))
-            return False
-    
-    def _set_aws_secret(self, key: str, value: str) -> bool:
-        """Set secret in AWS Secrets Manager"""
-        try:
-            secret_name = self.config.get("secret_name", "isa-model/secrets")
-            
-            # Get existing secrets
-            try:
-                response = self.secrets_client.get_secret_value(SecretId=secret_name)
-                existing_secrets = json.loads(response["SecretString"])
-            except:
-                existing_secrets = {}
-            
-            # Update with new secret
-            existing_secrets[key] = value
-            
-            # Update secret
-            self.secrets_client.update_secret(
-                SecretId=secret_name,
-                SecretString=json.dumps(existing_secrets)
-            )
-            
-            # Update cache
-            self._cache[key] = value
-            return True
-            
-        except Exception as e:
-            logger.error("Failed to set secret in AWS", key=key, error=str(e))
-            return False
-    
-    def list_secrets(self) -> List[str]:
-        """List available secret keys"""
-        try:
-            if self.backend == "vault":
-                return self._list_vault_secrets()
-            elif self.backend == "aws":
-                return self._list_aws_secrets()
-            elif self.backend == "env":
-                # Return common secret environment variables
-                common_secrets = [
-                    "OPENAI_API_KEY", "REPLICATE_API_TOKEN", "ANTHROPIC_API_KEY",
-                    "DATABASE_URL", "REDIS_URL", "ISA_API_KEY"
-                ]
-                return [key for key in common_secrets if os.getenv(key)]
-            else:
-                return []
-        except Exception as e:
-            logger.error("Failed to list secrets", error=str(e))
-            return []
-    
-    def _list_vault_secrets(self) -> List[str]:
-        """List secrets in HashiCorp Vault"""
-        try:
-            secret_path = self.config.get("secret_path", "secret/data/isa-model")
-            response = self.vault_client.secrets.kv.v2.read_secret_version(path=secret_path)
-            return list(response["data"]["data"].keys())
-        except Exception as e:
-            logger.warning("Failed to list Vault secrets", error=str(e))
-            return []
-    
-    def _list_aws_secrets(self) -> List[str]:
-        """List secrets in AWS Secrets Manager"""
-        try:
-            secret_name = self.config.get("secret_name", "isa-model/secrets")
-            response = self.secrets_client.get_secret_value(SecretId=secret_name)
-            secrets = json.loads(response["SecretString"])
-            return list(secrets.keys())
-        except Exception as e:
-            logger.warning("Failed to list AWS secrets", error=str(e))
-            return []
-    
-    def rotate_secret(self, key: str) -> bool:
-        """Rotate a secret (implementation depends on secret type)"""
-        # This is a placeholder for secret rotation logic
-        logger.info("Secret rotation requested", key=key)
-        # In production, this would implement proper rotation logic
-        return True
-    
-    def clear_cache(self):
-        """Clear the secrets cache"""
-        self._cache.clear()
-        logger.info("Secrets cache cleared")
-
-# Global secrets manager instance
-_secrets_manager = None
-
-def get_secrets_manager() -> SecretsManager:
-    """Get the global secrets manager instance"""
-    global _secrets_manager
-    
-    if _secrets_manager is None:
-        # Determine backend from environment
-        backend = os.getenv("SECRETS_BACKEND", "env")
-        
-        # Initialize with backend-specific configuration
-        if backend == "vault":
-            _secrets_manager = SecretsManager(
-                backend="vault",
-                vault_url=os.getenv("VAULT_URL"),
-                vault_token=os.getenv("VAULT_TOKEN"),
-                secret_path=os.getenv("VAULT_SECRET_PATH", "secret/data/isa-model")
-            )
-        elif backend == "aws":
-            _secrets_manager = SecretsManager(
-                backend="aws",
-                region=os.getenv("AWS_REGION", "us-east-1"),
-                secret_name=os.getenv("AWS_SECRET_NAME", "isa-model/secrets")
-            )
-        else:
-            _secrets_manager = SecretsManager(backend="env")
-    
-    return _secrets_manager
-
-def get_secret(key: str, default: Optional[str] = None) -> Optional[str]:
-    """Convenience function to get a secret"""
-    return get_secrets_manager().get_secret(key, default)
-
-def set_secret(key: str, value: str) -> bool:
-    """Convenience function to set a secret"""
-    return get_secrets_manager().set_secret(key, value)
-
-# Predefined secret getters for common secrets
-def get_openai_api_key() -> Optional[str]:
-    """Get OpenAI API key"""
-    return get_secret("OPENAI_API_KEY")
-
-def get_replicate_api_token() -> Optional[str]:
-    """Get Replicate API token"""
-    return get_secret("REPLICATE_API_TOKEN")
-
-def get_anthropic_api_key() -> Optional[str]:
-    """Get Anthropic API key"""
-    return get_secret("ANTHROPIC_API_KEY")
-
-def get_database_url() -> Optional[str]:
-    """Get database URL"""
-    return get_secret("DATABASE_URL")
-
-def get_redis_url() -> Optional[str]:
-    """Get Redis URL"""
-    return get_secret("REDIS_URL", "redis://localhost:6379")
-
-def get_isa_api_key() -> Optional[str]:
-    """Get ISA API key"""
-    return get_secret("ISA_API_KEY")
-
-# Health check for secrets manager
-async def check_secrets_health() -> Dict[str, Any]:
-    """Check secrets manager health"""
-    try:
-        manager = get_secrets_manager()
-        
-        # Test basic functionality
-        test_secret = manager.get_secret("HEALTH_CHECK_TEST", "test")
-        
-        return {
-            "secrets_manager": "ok",
-            "backend": manager.backend,
-            "cached_secrets": len(manager._cache),
-            "status": "healthy"
-        }
-    except Exception as e:
-        return {
-            "secrets_manager": "error",
-            "status": "unhealthy",
-            "error": str(e)
-        }