models

package
v0.10.0 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Jan 4, 2026 License: MIT Imports: 0 Imported by: 0

README

OmniLLM Models Catalog

This package provides a comprehensive catalog of LLM model identifiers and documentation references for all supported providers.

Purpose

  • Centralized Model Constants: All model IDs in one organized location
  • Documentation URLs: Official reference pages for each provider
  • Easy Updates: When providers release new models, update in one place
  • Type Safety: Use constants instead of hardcoded strings

Usage

Import the Package
import "github.com/agentplexus/omnillm/models"
Use Model Constants
// Anthropic Claude
client, _ := omnillm.NewClient(omnillm.ClientConfig{
    Provider: omnillm.ProviderNameAnthropic,
    APIKey:   apiKey,
})
response, _ := client.CreateChatCompletion(ctx, &omnillm.ChatCompletionRequest{
    Model: models.ClaudeOpus4_1,
    Messages: messages,
})

// OpenAI
response, _ := client.CreateChatCompletion(ctx, &omnillm.ChatCompletionRequest{
    Model: models.GPT4o,
    Messages: messages,
})

// X.AI Grok
response, _ := client.CreateChatCompletion(ctx, &omnillm.ChatCompletionRequest{
    Model: models.Grok4_1FastReasoning,
    Messages: messages,
})

// Google Gemini
response, _ := client.CreateChatCompletion(ctx, &omnillm.ChatCompletionRequest{
    Model: models.Gemini2_5Pro,
    Messages: messages,
})
Access Documentation URLs
// Print documentation URLs for reference
fmt.Println("Anthropic Models:", models.AnthropicModelsURL)
fmt.Println("OpenAI Models:", models.OpenAIModelsURL)
fmt.Println("X.AI Models:", models.XAIModelsURL)
fmt.Println("Gemini Models:", models.GeminiModelsURL)
fmt.Println("Bedrock Models:", models.BedrockModelsURL)
fmt.Println("Ollama Models:", models.OllamaModelsURL)

Package Structure

models/
├── doc.go          # Package documentation
├── README.md       # This file
├── anthropic.go    # Claude models + docs URL
├── openai.go       # OpenAI models + docs URL
├── xai.go          # X.AI Grok models + docs URL
├── gemini.go       # Google Gemini models + docs URL
├── bedrock.go      # AWS Bedrock models + docs URL
├── ollama.go       # Ollama models + docs URL
└── vertex.go       # Google Vertex AI models + docs URL

Updating Models

When providers release new models or deprecate existing ones:

  1. Check Documentation: Use the provider's ModelsURL constant to visit their docs
  2. Update Constants: Add new models or mark deprecated ones
  3. Update Constants Package: Update root constants.go if needed for backwards compatibility
  4. Update Tests: Update integration tests to use latest models
  5. Update Examples: Update example code to showcase new models
Example Update Workflow
# 1. Check X.AI documentation
open https://docs.x.ai/docs/models

# 2. Add new model constants to models/xai.go
# 3. Update constants.go for backwards compatibility
# 4. Update tests and examples
# 5. Update RELEASE_NOTES.md

Model Categories by Provider

Anthropic Claude
OpenAI
X.AI Grok
  • Latest: Grok 4.1 Fast (Reasoning/Non-Reasoning)
  • Current: Grok 4 (0709, Fast variants), Grok Code Fast
  • Previous: Grok 3, Grok 3 Mini
  • Legacy: Grok 2, Grok 2 Vision
  • Documentation: https://docs.x.ai/docs/models
Google Gemini
AWS Bedrock
Ollama (Local Models)
  • Llama: Llama 3 8B, Llama 3 70B
  • Mistral: Mistral 7B, Mixtral 8x7B
  • Code: CodeLlama, DeepSeek Coder
  • Other: Gemma, Qwen 2.5
  • Documentation: https://ollama.com/library

Benefits

1. Type Safety
// ✅ Good: Using constants (typo protection, autocomplete)
model := models.ClaudeOpus4_1

// ❌ Bad: Hardcoded strings (prone to typos)
model := "claude-opus-4-1-20250805"
2. Documentation
// Every model has inline comments explaining features
// Hover over constant in IDE to see documentation
models.Grok4_1FastReasoning  // Shows: "Best tool-calling model with 2M context"
3. Centralized Updates
// When a model ID changes or is deprecated, update in one place
// All code using the constant automatically uses new value
4. Easy Discovery
// IDE autocomplete shows all available models
models.Claude  // → Claude3Haiku, Claude3Opus, ClaudeOpus4, ClaudeOpus4_1, etc.
models.GPT     // → GPT35Turbo, GPT4Turbo, GPT4o, GPT4oMini, GPT5, etc.
models.Grok    // → Grok2_1212, Grok3, Grok3Mini, Grok4_0709, etc.

Compatibility

The root constants.go file maintains backwards compatibility by re-exporting models from this package:

// constants.go
const (
    // For backwards compatibility, these re-export from models package
    ModelClaudeOpus4 = models.ClaudeOpus4
    ModelGPT4o = models.GPT4o
    ModelGrok4_1FastReasoning = models.Grok4_1FastReasoning
)

Existing code continues to work, but new code should import models package directly for better organization.

Contributing

When adding support for a new provider:

  1. Create models/<provider>.go file
  2. Add <Provider>ModelsURL constant
  3. Add <Provider>APIURL constant
  4. Add model constants with documentation comments
  5. Update this README.md
  6. Update root constants.go for backwards compatibility
  7. Add tests if needed

License

MIT - See root LICENSE file

Documentation

Overview

Package models provides a comprehensive catalog of LLM model identifiers and documentation references for all supported providers.

This package organizes model constants by provider and includes reference URLs to official documentation pages for staying up-to-date with new models and changes.

Usage: 

import "github.com/agentplexus/omnillm/models"

// Use model constants
model := models.ClaudeOpus4
model := models.GPT4o
model := models.Grok4_1FastReasoning

// Reference documentation URLs for updates
fmt.Println(models.AnthropicModelsURL)
fmt.Println(models.OpenAIModelsURL)

Index

Constants

View Source 
const (
	// AnthropicModelsURL is the official Anthropic models documentation page.
	// Use this to check for new models, deprecations, and model updates.
	AnthropicModelsURL = "https://platform.claude.com/docs/en/about-claude/models/overview"

	// AnthropicAPIURL is the Anthropic API reference page.
	AnthropicAPIURL = "https://docs.anthropic.com/en/api"
)

Anthropic Claude Model Documentation

View Source 
const (
	ClaudeOpus4_5   = "claude-opus-4-5-20251101"   // Claude Opus 4.5 (November 2025)
	ClaudeSonnet4_5 = "claude-sonnet-4-5-20250929" // Claude Sonnet 4.5 (September 2025)
	ClaudeHaiku4_5  = "claude-haiku-4-5-20251001"  // Claude Haiku 4.5 (October 2025)
)

Claude 4.5 Family (Latest)

View Source 
const (
	ClaudeOpus4_1 = "claude-opus-4-1-20250805" // Claude Opus 4.1 (August 2025)
	ClaudeOpus4   = "claude-opus-4-20250514"   // Claude Opus 4 (May 2025)
)

Claude Opus 4 Family

View Source 
const (
	Claude3Opus   = "claude-3-opus-20240229"   // Claude 3 Opus (February 2024)
	Claude3Sonnet = "claude-3-sonnet-20240229" // Claude 3 Sonnet (February 2024)
	Claude3Haiku  = "claude-3-haiku-20240307"  // Claude 3 Haiku (March 2024)
)

Claude 3 Family

View Source 
const (
	// BedrockModelsURL is the official AWS Bedrock models documentation page.
	// Use this to check for new models, deprecations, and model updates.
	BedrockModelsURL = "https://docs.aws.amazon.com/bedrock/latest/userguide/models-supported.html"

	// BedrockAPIURL is the AWS Bedrock API reference page.
	BedrockAPIURL = "https://docs.aws.amazon.com/bedrock/latest/APIReference/welcome.html"
)

AWS Bedrock Model Documentation

View Source 
const (
	// BedrockClaudeOpus4 is Claude Opus 4 on AWS Bedrock.
	BedrockClaudeOpus4 = "anthropic.claude-opus-4-20250514-v1:0"

	// BedrockClaude3Opus is Claude 3 Opus on AWS Bedrock.
	BedrockClaude3Opus = "anthropic.claude-3-opus-20240229-v1:0"

	// BedrockClaude3Sonnet is Claude 3 Sonnet on AWS Bedrock.
	BedrockClaude3Sonnet = "anthropic.claude-3-sonnet-20240229-v1:0"
)

Bedrock Claude Models

View Source 
const (
	// GeminiModelsURL is the official Google Gemini models documentation page.
	// Use this to check for new models, deprecations, and model updates.
	GeminiModelsURL = "https://ai.google.dev/gemini-api/docs/models/gemini"

	// GeminiAPIURL is the Google Gemini API reference page.
	GeminiAPIURL = "https://ai.google.dev/gemini-api/docs"
)

Google Gemini Model Documentation

View Source 
const (
	// Gemini2_5Pro is stable with advanced reasoning capabilities.
	Gemini2_5Pro = "gemini-2.5-pro"

	// Gemini2_5Flash is stable with balanced performance.
	Gemini2_5Flash = "gemini-2.5-flash"

	// GeminiLive2_5Flash is the stable Live API model (private GA).
	GeminiLive2_5Flash = "gemini-live-2.5-flash"
)

Gemini 2.5 Family (Latest)

View Source 
const (
	Gemini1_5Pro   = "gemini-1.5-pro"   // Gemini 1.5 Pro
	Gemini1_5Flash = "gemini-1.5-flash" // Gemini 1.5 Flash
)

Gemini 1.5 Family

View Source 
const (
	// OllamaModelsURL is the official Ollama models library page.
	// Use this to check for new models and model updates.
	OllamaModelsURL = "https://ollama.com/library"

	// OllamaAPIURL is the Ollama API reference page.
	OllamaAPIURL = "https://github.com/ollama/ollama/blob/main/docs/api.md"
)

Ollama Model Documentation

View Source 
const (
	OllamaLlama3_8B  = "llama3:8b"  // Llama 3 8B
	OllamaLlama3_70B = "llama3:70b" // Llama 3 70B
)

Ollama Llama Models

View Source 
const (
	OllamaMistral7B   = "mistral:7b"   // Mistral 7B
	OllamaMixtral8x7B = "mixtral:8x7b" // Mixtral 8x7B
)

Ollama Mistral Models

View Source 
const (
	OllamaCodeLlama = "codellama:13b"       // CodeLlama 13B
	OllamaDeepSeek  = "deepseek-coder:6.7b" // DeepSeek Coder 6.7B
)

Ollama Code Models

View Source 
const (
	OllamaGemma2B = "gemma:2b" // Gemma 2B
	OllamaGemma7B = "gemma:7b" // Gemma 7B
)

Ollama Gemma Models

View Source 
const (
	// OpenAIModelsURL is the official OpenAI models documentation page.
	// Use this to check for new models, deprecations, and model updates.
	OpenAIModelsURL = "https://platform.openai.com/docs/models"

	// OpenAIAPIURL is the OpenAI API reference page.
	OpenAIAPIURL = "https://platform.openai.com/docs/api-reference"
)

OpenAI Model Documentation

View Source 
const (
	GPT5           = "gpt-5"             // GPT-5 latest
	GPT5Mini       = "gpt-5-mini"        // GPT-5 Mini
	GPT5Nano       = "gpt-5-nano"        // GPT-5 Nano
	GPT5ChatLatest = "gpt-5-chat-latest" // GPT-5 Chat Latest
)

GPT-5 Family (Latest)

View Source 
const (
	GPT4_1     = "gpt-4.1"      // GPT-4.1
	GPT4_1Mini = "gpt-4.1-mini" // GPT-4.1 Mini
	GPT4_1Nano = "gpt-4.1-nano" // GPT-4.1 Nano
)

GPT-4.1 Family

View Source 
const (
	GPT4o     = "gpt-4o"      // GPT-4o
	GPT4oMini = "gpt-4o-mini" // GPT-4o Mini
)

GPT-4o Family

View Source 
const (
	// VertexModelsURL is the official Google Vertex AI models documentation page.
	// Use this to check for new models, deprecations, and model updates.
	VertexModelsURL = "https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models"

	// VertexAPIURL is the Google Vertex AI API reference page.
	VertexAPIURL = "https://cloud.google.com/vertex-ai/generative-ai/docs/reference/rest"
)

Google Vertex AI Model Documentation

View Source 
const (
	// XAIModelsURL is the official X.AI models documentation page.
	// Use this to check for new models, deprecations, and model updates.
	XAIModelsURL = "https://docs.x.ai/docs/models"

	// XAIapiurl is the X.AI API reference page.
	XAIAPIURL = "https://docs.x.ai/docs"
)

X.AI Grok Model Documentation

View Source 
const (
	// Grok4_1FastReasoning is the best tool-calling model with 2M context window.
	// Optimized for maximum intelligence and agentic tool calling.
	Grok4_1FastReasoning = "grok-4-1-fast-reasoning"

	// Grok4_1FastNonReasoning provides instant responses with 2M context window.
	// Optimized for speed without reasoning overhead.
	Grok4_1FastNonReasoning = "grok-4-1-fast-non-reasoning"
)

Grok 4.1 Family (Latest - November 2025)

View Source 
const (
	// Grok4_0709 is the flagship Grok 4 model with 256K context window.
	// Released July 9, 2025. Provides high-quality reasoning.
	Grok4_0709 = "grok-4-0709"

	// Grok4FastReasoning provides fast reasoning with 2M context window.
	// Performance on par with grok-4-0709 with larger context.
	Grok4FastReasoning = "grok-4-fast-reasoning"

	// Grok4FastNonReasoning provides fast non-reasoning with 2M context window.
	// Optimized for speed without reasoning overhead.
	Grok4FastNonReasoning = "grok-4-fast-non-reasoning"

	// GrokCodeFast1 is optimized for agentic coding with 256K context window.
	// Speedy and economical reasoning model for coding tasks.
	GrokCodeFast1 = "grok-code-fast-1"
)

Grok 4 Family (July 2025)

View Source 
const (
	Grok3     = "grok-3"      // Grok 3
	Grok3Mini = "grok-3-mini" // Grok 3 Mini (smaller, faster)
)

Grok 3 Family

View Source 
const (
	Grok2_1212   = "grok-2-1212"        // Grok 2 (December 2024)
	Grok2_Vision = "grok-2-vision-1212" // Grok 2 with vision capabilities
)

Grok 2 Family

View Source 
const (
	// GrokBeta is deprecated. Use Grok3 or Grok4_1FastReasoning instead.
	GrokBeta = "grok-beta" // Deprecated: use grok-3 or grok-4

	// GrokVision is deprecated. Use Grok2_Vision instead.
	GrokVision = "grok-vision-beta" // Deprecated
)

Deprecated Grok Models

View Source 
const (
	// BedrockTitan is Amazon Titan Text Express on AWS Bedrock.
	BedrockTitan = "amazon.titan-text-express-v1"
)

Bedrock Amazon Titan Models

View Source 
const (
	Claude3_5Haiku = "claude-3-5-haiku-20241022" // Claude 3.5 Haiku (October 2024)
)

Claude 3.5 Family

View Source 
const (
	Claude3_7Sonnet = "claude-3-7-sonnet-20250219" // Claude 3.7 Sonnet (February 2025)
)

Claude 3.7 Sonnet

View Source 
const (
	ClaudeSonnet4 = "claude-sonnet-4-20250514" // Claude Sonnet 4 (May 2025)
)

Claude Sonnet 4 Family

View Source 
const (
	GPT35Turbo = "gpt-3.5-turbo" // GPT-3.5 Turbo
)

GPT-3.5 Family

View Source 
const (
	GPT4Turbo = "gpt-4-turbo" // GPT-4 Turbo
)

GPT-4 Family

View Source 
const (
	GeminiPro = "gemini-pro" // Legacy Gemini Pro (use Gemini 2.5 instead)
)

Legacy Gemini Models

View Source 
const (
	OllamaQwen2_5 = "qwen2.5:7b" // Qwen 2.5 7B
)

Ollama Other Models

View Source 
const (
	// VertexClaudeOpus4 is Claude Opus 4 on Google Vertex AI.
	VertexClaudeOpus4 = "claude-opus-4@20250514"
)

Vertex AI Claude Models

Variables

This section is empty.

Functions

This section is empty.

Types

This section is empty.

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.