haive.core.models.llm.providers.anthropic¶

Anthropic Provider Module.

This module implements the Anthropic language model provider for the Haive framework, supporting Claude 3 models (Opus, Sonnet, Haiku) and earlier Claude versions.

The provider handles API key management, model configuration, and safe imports of the langchain-anthropic package dependencies.

Examples

Basic usage:

from haive.core.models.llm.providers.anthropic import AnthropicProvider

provider = AnthropicProvider(
    model="claude-3-opus-20240229",
    temperature=0.7,
    max_tokens=4096
)
llm = provider.instantiate()

With streaming:

provider = AnthropicProvider(
    model="claude-3-sonnet-20240229",
    streaming=True
)

async for chunk in provider.instantiate().astream("Tell me a story"):
    print(chunk.content, end="")

AnthropicProvider(*[, requests_per_second, ...])

Anthropic language model provider configuration.

Classes¶

AnthropicProvider

Anthropic language model provider configuration.

Module Contents¶

class haive.core.models.llm.providers.anthropic.AnthropicProvider(/, **data)[source]¶

Bases: haive.core.models.llm.providers.base.BaseLLMProvider

Anthropic language model provider configuration.

This provider supports all Anthropic Claude models including Claude 3 (Opus, Sonnet, Haiku) and Claude 2 variants. It provides access to Anthropic’s constitutional AI models with support for large context windows.

Parameters:

  • data (Any)

  • requests_per_second (float | None)

  • tokens_per_second (int | None)

  • tokens_per_minute (int | None)

  • max_retries (int)

  • retry_delay (float)

  • check_every_n_seconds (float | None)

  • burst_size (int | None)

  • provider (LLMProvider)

  • model (str | None)

  • name (str | None)

  • api_key (SecretStr)

  • cache_enabled (bool)

  • cache_ttl (int | None)

  • extra_params (dict[str, Any] | None)

  • debug (bool)

  • temperature (float | None)

  • max_tokens (int | None)

  • top_p (float | None)

  • top_k (int | None)

  • streaming (bool)

provider¶

Always LLMProvider.ANTHROPIC

model¶

Model name (default: “claude-3-sonnet-20240229”)

temperature¶

Sampling temperature (0-1)

max_tokens¶

Maximum tokens to generate (up to 4096)

top_p¶

Nucleus sampling parameter

top_k¶

Top-k sampling parameter

streaming¶

Whether to stream responses

Environment Variables:

ANTHROPIC_API_KEY: API key for authentication ANTHROPIC_MODEL: Default model to use

Model Variants:

  • claude-3-opus-20240229: Most capable, best for complex tasks

  • claude-3-sonnet-20240229: Balanced performance and speed

  • claude-3-haiku-20240307: Fastest, most cost-effective

  • claude-2.1: Previous generation, 200K context

  • claude-2.0: Previous generation, 100K context

Examples

Using Claude 3 Opus:

provider = AnthropicProvider(
    model="claude-3-opus-20240229",
    temperature=0.5,
    max_tokens=4096
)
llm = provider.instantiate()

With custom sampling:

provider = AnthropicProvider(
    model="claude-3-sonnet-20240229",
    temperature=0.8,
    top_p=0.9,
    top_k=40
)

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

classmethod get_models()[source]¶

Get available Anthropic models.

Note: Anthropic doesn’t provide a public API to list models, so this returns a hardcoded list of known models.

Returns:

List of available model names

Return type:

list[str]

max_tokens: int | None = None¶

Get maximum total tokens for this model.