isa-model 0.3.91__py3-none-any.whl → 0.4.3__py3-none-any.whl

isa_model/deployment/models/org-org-acme-corp-tenant-a-service-llm-20250825-225822/tenant-a-service_modal_service.py

@@ -0,0 +1,48 @@
+"""
+tenant-a-service LLM Service for Modal
+
+Auto-generated service for model: gpt2
+Architecture: gpt
+"""
+
+import modal
+from typing import Dict, Any, List
+
+app = modal.App("tenant-a-service")
+
+image = modal.Image.debian_slim().pip_install(
+    "accelerate>=0.24.0", "transformers>=4.35.0", "httpx>=0.26.0", "torch>=2.0.0", "requests>=2.31.0", "numpy>=1.24.0", "pydantic>=2.0.0"
+)
+
+@app.cls(
+    image=image,
+    gpu=modal.gpu.A10G(count=1),
+    container_idle_timeout=300,
+    memory=32768
+)
+class Tenant_A_ServiceService:
+    
+    @modal.enter()
+    def load_model(self):
+        import torch
+        from transformers import AutoTokenizer, AutoModelForCausalLM
+        
+        self.tokenizer = AutoTokenizer.from_pretrained("gpt2")
+        self.model = AutoModelForCausalLM.from_pretrained(
+            "gpt2",
+            torch_dtype=torch.float16,
+            device_map="auto",
+            trust_remote_code=True
+        )
+    
+    @modal.method()
+    def generate(self, messages: List[Dict[str, str]], **kwargs):
+        # Generate response (simplified)
+        prompt = messages[-1]["content"] if messages else ""
+        return {"response": f"Generated response for: {prompt}", "model": "gpt2"}
+
+@app.function(image=image)
+@modal.web_endpoint(method="POST")
+def inference_endpoint(item: Dict[str, Any]):
+    service = Tenant_A_ServiceService()
+    return service.generate(**item)

isa_model/deployment/models/org-test-org-123-prefix-test-service-llm-20250825-225822/prefix-test-service_modal_service.py

@@ -0,0 +1,48 @@
+"""
+prefix-test-service LLM Service for Modal
+
+Auto-generated service for model: gpt2
+Architecture: gpt
+"""
+
+import modal
+from typing import Dict, Any, List
+
+app = modal.App("prefix-test-service")
+
+image = modal.Image.debian_slim().pip_install(
+    "accelerate>=0.24.0", "transformers>=4.35.0", "httpx>=0.26.0", "torch>=2.0.0", "requests>=2.31.0", "numpy>=1.24.0", "pydantic>=2.0.0"
+)
+
+@app.cls(
+    image=image,
+    gpu=modal.gpu.A10G(count=1),
+    container_idle_timeout=300,
+    memory=32768
+)
+class Prefix_Test_ServiceService:
+    
+    @modal.enter()
+    def load_model(self):
+        import torch
+        from transformers import AutoTokenizer, AutoModelForCausalLM
+        
+        self.tokenizer = AutoTokenizer.from_pretrained("gpt2")
+        self.model = AutoModelForCausalLM.from_pretrained(
+            "gpt2",
+            torch_dtype=torch.float16,
+            device_map="auto",
+            trust_remote_code=True
+        )
+    
+    @modal.method()
+    def generate(self, messages: List[Dict[str, str]], **kwargs):
+        # Generate response (simplified)
+        prompt = messages[-1]["content"] if messages else ""
+        return {"response": f"Generated response for: {prompt}", "model": "gpt2"}
+
+@app.function(image=image)
+@modal.web_endpoint(method="POST")
+def inference_endpoint(item: Dict[str, Any]):
+    service = Prefix_Test_ServiceService()
+    return service.generate(**item)

isa_model/deployment/models/test-llm-service-llm-20250825-204442/test-llm-service_modal_service.py

@@ -0,0 +1,48 @@
+"""
+test-llm-service LLM Service for Modal
+
+Auto-generated service for model: gpt2
+Architecture: gpt
+"""
+
+import modal
+from typing import Dict, Any, List
+
+app = modal.App("test-llm-service")
+
+image = modal.Image.debian_slim().pip_install(
+    "torch>=2.0.0", "httpx>=0.26.0", "transformers>=4.35.0", "requests>=2.31.0", "pydantic>=2.0.0", "numpy>=1.24.0", "accelerate>=0.24.0"
+)
+
+@app.cls(
+    image=image,
+    gpu=modal.gpu.A10G(count=1),
+    container_idle_timeout=300,
+    memory=32768
+)
+class Test_Llm_ServiceService:
+    
+    @modal.enter()
+    def load_model(self):
+        import torch
+        from transformers import AutoTokenizer, AutoModelForCausalLM
+        
+        self.tokenizer = AutoTokenizer.from_pretrained("gpt2")
+        self.model = AutoModelForCausalLM.from_pretrained(
+            "gpt2",
+            torch_dtype=torch.float16,
+            device_map="auto",
+            trust_remote_code=True
+        )
+    
+    @modal.method()
+    def generate(self, messages: List[Dict[str, str]], **kwargs):
+        # Generate response (simplified)
+        prompt = messages[-1]["content"] if messages else ""
+        return {"response": f"Generated response for: {prompt}", "model": "gpt2"}
+
+@app.function(image=image)
+@modal.web_endpoint(method="POST")
+def inference_endpoint(item: Dict[str, Any]):
+    service = Test_Llm_ServiceService()
+    return service.generate(**item)

isa_model/deployment/models/test-monitoring-gpt2-llm-20250825-212906/test-monitoring-gpt2_modal_service.py

@@ -0,0 +1,48 @@
+"""
+test-monitoring-gpt2 LLM Service for Modal
+
+Auto-generated service for model: gpt2
+Architecture: gpt
+"""
+
+import modal
+from typing import Dict, Any, List
+
+app = modal.App("test-monitoring-gpt2")
+
+image = modal.Image.debian_slim().pip_install(
+    "numpy>=1.24.0", "requests>=2.31.0", "accelerate>=0.24.0", "httpx>=0.26.0", "pydantic>=2.0.0", "transformers>=4.35.0", "torch>=2.0.0"
+)
+
+@app.cls(
+    image=image,
+    gpu=modal.gpu.A10G(count=1),
+    container_idle_timeout=300,
+    memory=32768
+)
+class Test_Monitoring_Gpt2Service:
+    
+    @modal.enter()
+    def load_model(self):
+        import torch
+        from transformers import AutoTokenizer, AutoModelForCausalLM
+        
+        self.tokenizer = AutoTokenizer.from_pretrained("gpt2")
+        self.model = AutoModelForCausalLM.from_pretrained(
+            "gpt2",
+            torch_dtype=torch.float16,
+            device_map="auto",
+            trust_remote_code=True
+        )
+    
+    @modal.method()
+    def generate(self, messages: List[Dict[str, str]], **kwargs):
+        # Generate response (simplified)
+        prompt = messages[-1]["content"] if messages else ""
+        return {"response": f"Generated response for: {prompt}", "model": "gpt2"}
+
+@app.function(image=image)
+@modal.web_endpoint(method="POST")
+def inference_endpoint(item: Dict[str, Any]):
+    service = Test_Monitoring_Gpt2Service()
+    return service.generate(**item)

isa_model/deployment/models/test-monitoring-gpt2-llm-20250825-213009/test-monitoring-gpt2_modal_service.py

@@ -0,0 +1,48 @@
+"""
+test-monitoring-gpt2 LLM Service for Modal
+
+Auto-generated service for model: gpt2
+Architecture: gpt
+"""
+
+import modal
+from typing import Dict, Any, List
+
+app = modal.App("test-monitoring-gpt2")
+
+image = modal.Image.debian_slim().pip_install(
+    "transformers>=4.35.0", "torch>=2.0.0", "accelerate>=0.24.0", "httpx>=0.26.0", "numpy>=1.24.0", "requests>=2.31.0", "pydantic>=2.0.0"
+)
+
+@app.cls(
+    image=image,
+    gpu=modal.gpu.A10G(count=1),
+    container_idle_timeout=300,
+    memory=32768
+)
+class Test_Monitoring_Gpt2Service:
+    
+    @modal.enter()
+    def load_model(self):
+        import torch
+        from transformers import AutoTokenizer, AutoModelForCausalLM
+        
+        self.tokenizer = AutoTokenizer.from_pretrained("gpt2")
+        self.model = AutoModelForCausalLM.from_pretrained(
+            "gpt2",
+            torch_dtype=torch.float16,
+            device_map="auto",
+            trust_remote_code=True
+        )
+    
+    @modal.method()
+    def generate(self, messages: List[Dict[str, str]], **kwargs):
+        # Generate response (simplified)
+        prompt = messages[-1]["content"] if messages else ""
+        return {"response": f"Generated response for: {prompt}", "model": "gpt2"}
+
+@app.function(image=image)
+@modal.web_endpoint(method="POST")
+def inference_endpoint(item: Dict[str, Any]):
+    service = Test_Monitoring_Gpt2Service()
+    return service.generate(**item)

isa_model/deployment/storage/__init__.py

@@ -0,0 +1,5 @@
+"""Storage and persistence for deployments"""
+
+from .deployment_repository import DeploymentRepository
+
+__all__ = ["DeploymentRepository"]