PyPI - llmstudio-core - Versions diffs - 0.0.1a1__py3-none-any.whl - Mend

llmstudio-core 0.0.1a1__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (12) hide show

llmstudio_core/__init__.py +0 -0
llmstudio_core/config.yaml +374 -0
llmstudio_core/exceptions.py +6 -0
llmstudio_core/providers/__init__.py +35 -0
llmstudio_core/providers/azure.py +601 -0
llmstudio_core/providers/openai.py +70 -0
llmstudio_core/providers/provider.py +676 -0
llmstudio_core/providers/vertexai.py +569 -0
llmstudio_core/utils.py +65 -0
llmstudio_core-0.0.1a1.dist-info/METADATA +79 -0
llmstudio_core-0.0.1a1.dist-info/RECORD +12 -0
llmstudio_core-0.0.1a1.dist-info/WHEEL +4 -0

llmstudio_core/__init__.py ADDED Viewed

File without changes

llmstudio_core/config.yaml ADDED Viewed

@@ -0,0 +1,374 @@
+providers:
+  anthropic:
+    id: anthropic
+    name: Anthropic
+    chat: true
+    embed: true
+    keys:
+      - ANTHROPIC_API_KEY
+    models:
+      claude-3-opus-20240229:
+        mode: chat
+        max_tokens: 200000
+        input_token_cost: 0.000015
+        output_token_cost: 0.000075
+      claude-3-sonnet-2024022:
+        mode: chat
+        max_tokens: 200000
+        input_token_cost: 0.000003
+        output_token_cost: 0.000015
+      claude-3-haiku-20240307:
+        mode: chat
+        max_tokens: 200000
+        input_token_cost: 0.00000025
+        output_token_cost: 0.00000125
+      claude-2.1:
+        mode: chat
+        max_tokens: 200000
+        input_token_cost: 0.000008
+        output_token_cost: 0.000024
+      claude-2:
+        mode: chat
+        max_tokens: 100000
+        input_token_cost: 0.000008
+        output_token_cost: 0.000024
+      claude-instant-1.2:
+        mode: chat
+        max_tokens: 100000
+        input_token_cost: 0.00000163
+        output_token_cost: 0.00000551
+    parameters:
+      temperature:
+        name: "Temperature"
+        type: float
+        default: 1
+        min: 0
+        max: 1
+        step: 0.01
+      max_tokens:
+        name: "Maximum tokens"
+        type: float
+        default: 256
+        min: 1
+        max: 4096
+        step: 0.01
+      top_p:
+        name: "Top P"
+        type: float
+        default: 1
+        min: 0
+        max: 1
+        step: 0.01
+      top_k:
+        name: "Top K"
+        type: float
+        default: 5
+        min: 0
+        max: 500
+        step: 1
+  ollama:
+    id: ollama
+    name: Ollama
+    chat: true
+    embed: true
+    keys:
+    models:
+      llama2:
+        mode: chat
+        max_tokens: 0
+        input_token_cost: 0
+        output_token_cost: 0
+    parameters:
+      temperature:
+        name: "Temperature"
+        type: float
+        default: 1
+        min: 0
+        max: 1
+        step: 0.01
+      max_tokens:
+        name: "Maximum tokens"
+        type: float
+        default: 256
+        min: 1
+        max: 4096
+        step: 0.01
+      top_p:
+        name: "Top P"
+        type: float
+        default: 1
+        min: 0
+        max: 1
+        step: 0.01
+      top_k:
+        name: "Top K"
+        type: float
+        default: 5
+        min: 0
+        max: 500
+        step: 1
+  openai:
+    id: openai
+    name: OpenAI
+    chat: true
+    embed: true
+    keys:
+      - OPENAI_API_KEY
+    models:
+      gpt-4o-mini:
+        mode: chat
+        max_tokens: 128000
+        input_token_cost: 0.00000015
+        output_token_cost: 0.00000060
+      gpt-4o:
+        mode: chat
+        max_tokens: 128000
+        input_token_cost: 0.000005
+        output_token_cost: 0.000015
+      gpt-4-turbo:
+        mode: chat
+        max_tokens: 128000
+        input_token_cost: 0.00001
+        output_token_cost: 0.00003
+      gpt-4:
+        mode: chat
+        max_tokens: 8192
+        input_token_cost: 0.00003
+        output_token_cost: 0.00006
+      gpt-3.5-turbo:
+        mode: chat
+        max_tokens: 16385
+        input_token_cost: 0.0000005
+        output_token_cost: 0.0000015
+      gpt-3.5-turbo-instruct:
+        mode: chat
+        max_tokens: 4096
+        input_token_cost: 0.0000015
+        output_token_cost: 0.000002
+    parameters:
+      temperature:
+        name: "Temperature"
+        type: float
+        default: 1
+        min: 0
+        max: 2
+        step: 0.01
+      max_tokens:
+        name: "Maximum length"
+        type: int
+        default: 256
+        min: 1
+        max: 4096
+        step: 1
+      top_p:
+        name: "Top P"
+        type: float
+        default: 1
+        min: 0
+        max: 1
+        step: 0.01
+      frequency_penalty:
+        name: "Frequency Penalty"
+        type: float
+        default: 0
+        min: 0
+        max: 2
+        step: 0.01
+      presence_penalty:
+        name: "Presence Penalty"
+        type: float
+        default: 0
+        min: 0
+        max: 2
+        step: 0.01
+  azure:
+    id: azure
+    name: Azure
+    chat: true
+    embed: true
+    keys:
+      - AZURE_API_KEY
+      - AZURE_API_ENDPOINT
+      - AZURE_API_VERSION
+    models:
+      gpt-4o-mini:
+        mode: chat
+        max_tokens: 128000
+        input_token_cost: 0.00000015
+        output_token_cost: 0.0000006
+      gpt-4o:
+        mode: chat
+        max_tokens: 128000
+        input_token_cost: 0.000005
+        output_token_cost: 0.000015
+      gpt-4-turbo:
+        mode: chat
+        max_tokens: 128000
+        input_token_cost: 0.00001
+        output_token_cost: 0.00003
+      gpt-4:
+        mode: chat
+        max_tokens: 8192
+        input_token_cost: 0.00003
+        output_token_cost: 0.00006
+      gpt-35-turbo:
+        mode: chat
+        max_tokens: 4097
+        input_token_cost: 0.0000005
+        output_token_cost: 0.0000015
+      Meta-Llama-3.1-405B-Instruct:
+        mode: chat
+        max_tokens: 128000
+        input_token_cost: 0.00533
+        output_token_cost: 0.016
+      Meta-Llama-3.1-70B-Instruct:
+        mode: chat
+        max_tokens: 128000
+        input_token_cost: 0.00268
+        output_token_cost: 0.00354
+      Meta-Llama-3.1-8B-Instruct:
+        mode: chat
+        max_tokens: 128000
+        input_token_cost: 0.0003
+        output_token_cost: 0.00061
+      Meta-Llama-3-70B-Instruct:
+        mode: chat
+        max_tokens: 8192
+        input_token_cost: 0.00378
+        output_token_cost: 0.01134
+      Meta-Llama-3-8B-Instruct:
+        mode: chat
+        max_tokens: 8192
+        input_token_cost: 0.00037
+        output_token_cost: 0.0011
+      Meta-Llama-2-70B:
+        mode: chat
+        max_tokens: 4096
+        input_token_cost: 0.00154
+        output_token_cost: 0.00177
+      Meta-Llama-2-70B-Chat:
+        mode: chat
+        max_tokens: 4096
+        input_token_cost: 0.00154
+        output_token_cost: 0.00177
+      Meta-Llama-2-13B-Chat:
+        mode: chat
+        max_tokens: 4096
+        input_token_cost: 0.00081
+        output_token_cost: 0.00094
+      Meta-Llama-2-7B:
+        mode: chat
+        max_tokens: 4096
+        input_token_cost: 0.00052
+        output_token_cost: 0.00067
+    parameters:
+      temperature:
+        name: "Temperature"
+        type: float
+        default: 1
+        min: 0
+        max: 2
+        step: 0.01
+      max_tokens:
+        name: "Maximum length"
+        type: int
+        default: 256
+        min: 1
+        max: 4096
+        step: 1
+      top_p:
+        name: "Top P"
+        type: float
+        default: 1
+        min: 0
+        max: 1
+        step: 0.01
+      frequency_penalty:
+        name: "Frequency Penalty"
+        type: float
+        default: 0
+        min: 0
+        max: 2
+        step: 0.01
+      presence_penalty:
+        name: "Presence Penalty"
+        type: float
+        default: 0
+        min: 0
+        max: 2
+        step: 0.01
+  vertexai:
+    id: vertexai
+    name: VertexAI
+    chat: true
+    embed: true
+    keys:
+      - GOOGLE_API_KEY
+    models:
+      gemini-1.5-flash:
+        mode: chat
+        max_tokens: 1000000
+        input_token_cost:
+          - range: [0, 128000]
+            cost: 0.00000035
+          - range: [128001, null]
+            cost: 0.0000007
+        output_token_cost:
+          - range: [0, 128000]
+            cost: 0.00000105
+          - range: [128001, null]
+            cost: 0.0000021
+      gemini-1.5-pro-latest:
+        mode: chat
+        max_tokens: 1000000
+        input_token_cost:
+          - range: [0, 128000]
+            cost: 0.00000035
+          - range: [128001, null]
+            cost: 0.0000007
+        output_token_cost:
+          - range: [0, 128000]
+            cost: 0.0000105
+          - range: [128001, null]
+            cost: 0.000021
+      gemini-1.0-pro:
+        mode: chat
+        max_tokens: 1000000
+        input_token_cost: 0.0000005
+        output_token_cost: 0.0000015
+    parameters:
+      temperature:
+        name: "Temperature"
+        type: float
+        default: 1
+        min: 0
+        max: 2
+        step: 0.01
+      max_tokens:
+        name: "Maximum length"
+        type: int
+        default: 256
+        min: 1
+        max: 4096
+        step: 1
+      top_p:
+        name: "Top P"
+        type: float
+        default: 1
+        min: 0
+        max: 1
+        step: 0.01
+      frequency_penalty:
+        name: "Frequency Penalty"
+        type: float
+        default: 0
+        min: 0
+        max: 2
+        step: 0.01
+      presence_penalty:
+        name: "Presence Penalty"
+        type: float
+        default: 0
+        min: 0
+        max: 2
+        step: 0.01

llmstudio_core/exceptions.py ADDED Viewed

@@ -0,0 +1,6 @@
+class CustomError(Exception):
+    """Base class for all custom exceptions"""
+class ProviderError(CustomError):
+    """Raised when there is a validation error"""

llmstudio_core/providers/__init__.py ADDED Viewed

@@ -0,0 +1,35 @@
+# from llmstudio_core.providers.anthropic import AnthropicProvider #TODO: adpat it
+from typing import Optional
+from llmstudio_core.providers.azure import AzureProvider
+# from llmstudio_core.providers.ollama import OllamaProvider #TODO: adapt it
+from llmstudio_core.providers.openai import OpenAIProvider
+from llmstudio_core.providers.provider import ProviderCore, provider_registry
+from llmstudio_core.providers.vertexai import VertexAIProvider
+from llmstudio_core.utils import _load_providers_config
+_engine_config = _load_providers_config()
+def LLMCore(provider: str, api_key: Optional[str] = None, **kwargs) -> ProviderCore:
+    """
+    Factory method to create an instance of a provider.
+    Args:
+        provider (str): The name of the provider.
+        api_key (Optional[str], optional): The API key for the provider. Defaults to None.
+    Returns:
+        ProviderCore: An instance of the provider.
+    Raises:
+        NotImplementedError: If the provider is not found in the provider map.
+    """
+    provider_config = _engine_config.providers.get(provider)
+    provider_class = provider_registry.get(provider_config.id)
+    if provider_class:
+        return provider_class(config=provider_config, api_key=api_key, **kwargs)
+    raise NotImplementedError(
+        f"Provider not found: {provider_config.id}. Available providers: {str(provider_registry.keys())}"
+    )