edsl 0.1.34.dev2__py3-none-any.whl → 0.1.36__py3-none-any.whl

edsl/inference_services/InferenceServicesCollection.py

@@ -56,13 +56,19 @@ class InferenceServicesCollection:
         self.services.append(service)
 
     def create_model_factory(self, model_name: str, service_name=None, index=None):
+        from edsl.inference_services.TestService import TestService
+
+        if model_name == "test":
+            return TestService.create_model(model_name)
+
+        if service_name:
+            for service in self.services:
+                if service_name == service._inference_service_:
+                    return service.create_model(model_name)
+
         for service in self.services:
             if model_name in self._get_service_available(service):
                 if service_name is None or service_name == service._inference_service_:
                     return service.create_model(model_name)
 
-        # if model_name == "test":
-        #     from edsl.language_models import LanguageModel
-        #     return LanguageModel(test = True)
-
         raise Exception(f"Model {model_name} not found in any of the services")

edsl/inference_services/OpenAIService.py

@@ -188,12 +188,16 @@ class OpenAIService(InferenceServiceABC):
                 else:
                     content = user_prompt
                 client = self.async_client()
+
                 messages = [
                     {"role": "system", "content": system_prompt},
                     {"role": "user", "content": content},
                 ]
-                if system_prompt == "" and self.omit_system_prompt_if_empty:
+                if (
+                    system_prompt == "" and self.omit_system_prompt_if_empty
+                ) or "o1" in self.model:
                     messages = messages[1:]
+
                 params = {
                     "model": self.model,
                     "messages": messages,
@@ -205,7 +209,14 @@ class OpenAIService(InferenceServiceABC):
                     "logprobs": self.logprobs,
                     "top_logprobs": self.top_logprobs if self.logprobs else None,
                 }
-                response = await client.chat.completions.create(**params)
+                if "o1" in self.model:
+                    params.pop("max_tokens")
+                    params["max_completion_tokens"] = self.max_tokens
+                    params["temperature"] = 1
+                try:
+                    response = await client.chat.completions.create(**params)
+                except Exception as e:
+                    print(e)
                 return response.model_dump()
 
         LLM.__name__ = "LanguageModel"

edsl/inference_services/models_available_cache.py

@@ -65,7 +65,31 @@ models_available = {
         "meta-llama/Meta-Llama-3-70B-Instruct",
         "openchat/openchat_3.5",
     ],
-    "google": ["gemini-pro"],
+    "google": [
+        "gemini-1.0-pro",
+        "gemini-1.0-pro-001",
+        "gemini-1.0-pro-latest",
+        "gemini-1.0-pro-vision-latest",
+        "gemini-1.5-flash",
+        "gemini-1.5-flash-001",
+        "gemini-1.5-flash-001-tuning",
+        "gemini-1.5-flash-002",
+        "gemini-1.5-flash-8b",
+        "gemini-1.5-flash-8b-001",
+        "gemini-1.5-flash-8b-exp-0827",
+        "gemini-1.5-flash-8b-exp-0924",
+        "gemini-1.5-flash-8b-latest",
+        "gemini-1.5-flash-exp-0827",
+        "gemini-1.5-flash-latest",
+        "gemini-1.5-pro",
+        "gemini-1.5-pro-001",
+        "gemini-1.5-pro-002",
+        "gemini-1.5-pro-exp-0801",
+        "gemini-1.5-pro-exp-0827",
+        "gemini-1.5-pro-latest",
+        "gemini-pro",
+        "gemini-pro-vision",
+    ],
     "bedrock": [
         "amazon.titan-tg1-large",
         "amazon.titan-text-lite-v1",

edsl/inference_services/registry.py

@@ -11,21 +11,29 @@ from edsl.inference_services.AwsBedrock import AwsBedrockService
 from edsl.inference_services.AzureAI import AzureAIService
 from edsl.inference_services.OllamaService import OllamaService
 from edsl.inference_services.TestService import TestService
-from edsl.inference_services.MistralAIService import MistralAIService
 from edsl.inference_services.TogetherAIService import TogetherAIService
 
-default = InferenceServicesCollection(
-    [
-        OpenAIService,
-        AnthropicService,
-        DeepInfraService,
-        GoogleService,
-        GroqService,
-        AwsBedrockService,
-        AzureAIService,
-        OllamaService,
-        TestService,
-        MistralAIService,
-        TogetherAIService,
-    ]
-)
+try:
+    from edsl.inference_services.MistralAIService import MistralAIService
+
+    mistral_available = True
+except Exception as e:
+    mistral_available = False
+
+services = [
+    OpenAIService,
+    AnthropicService,
+    DeepInfraService,
+    GoogleService,
+    GroqService,
+    AwsBedrockService,
+    AzureAIService,
+    OllamaService,
+    TestService,
+    TogetherAIService,
+]
+
+if mistral_available:
+    services.append(MistralAIService)
+
+default = InferenceServicesCollection(services)