Crate serdes_ai_providers 

Provider abstractions for serdes-ai.

This crate provides a unified interface for different AI providers:

• OpenAI - GPT-4o, GPT-4 Turbo, o1, o3-mini
• Anthropic - Claude 3.5 Sonnet, Claude 3 Opus
• Google - Gemini 2.0 Flash, Gemini Pro
• Azure - Azure OpenAI Service
• Groq - Ultra-fast Llama, Mixtral
• Mistral - Mistral Large, Codestral
• Ollama - Local models
• Together AI - Open models
• Fireworks - Fast inference
• DeepSeek - DeepSeek Chat, DeepSeek R1
• OpenRouter - Multi-provider routing
• Cohere - Command R+
• Gateway - AI gateways (Portkey, LiteLLM, Helicone, Cloudflare)

Example

use serdes_ai_providers::{ProviderRegistry, OpenAIProvider};
use std::sync::Arc;

// Create a registry
let registry = ProviderRegistry::new();

// Register providers
registry.register(Arc::new(OpenAIProvider::new("sk-...")));

// Infer provider from model string
let (provider, model) = registry.infer_provider("openai:gpt-4o")?;

Model Strings

Models can be specified with provider prefixes:

• openai:gpt-4o - OpenAI GPT-4o
• anthropic:claude-3-5-sonnet-20241022 - Anthropic Claude
• google:gemini-2.0-flash - Google Gemini
• groq:llama-3.3-70b-versatile - Groq Llama
• ollama:llama3.2 - Local Ollama model

Or models can be inferred from their names:

• gpt-4o → OpenAI
• claude-3-5-sonnet-20241022 → Anthropic
• gemini-2.0-flash → Google

Re-exports

pub use oauth::config::chatgpt_oauth_config;

pub use oauth::config::claude_code_oauth_config;

pub use oauth::refresh_token;

pub use oauth::run_pkce_flow;

pub use oauth::OAuthConfig;

pub use oauth::OAuthContext;

pub use oauth::OAuthError;

pub use oauth::TokenResponse;

Modules

oauth

OAuth utilities for PKCE-based authentication flows.

prelude

Prelude for common imports.

Structs

AnthropicProvider

Anthropic provider.

CohereProvider

Cohere provider.

DeepSeekProvider

DeepSeek provider.

FireworksProvider

Fireworks AI provider.

GatewayConfig

Configuration for a gateway provider.

GatewayProvider

Gateway provider for AI routing services.

GoogleProvider

Google AI (Generative Language API) provider.

OpenAIProvider

OpenAI provider.

OpenRouterProvider

OpenRouter provider.

ProviderConfig

Common configuration for providers.

ProviderRegistry

Registry for looking up providers by name.

TogetherProvider

Together AI provider.

VertexAIProvider

Vertex AI provider.

Enums

ProviderError

Provider error types.

Traits

Provider

Provider trait - provides authenticated access to an AI API.

Functions

from_env

Create a provider registry configured from environment variables.

global_registry

Get the global provider registry.

infer

Infer provider and model from a model string.

infer_provider_from_model_name

Infer provider from model name.

Type Aliases

BoxedProvider

Type alias for boxed providers.