Supported Models

The Lunar SDK supports 100+ models from 11 providers. You can call any model directly (the API chooses the best provider) or force a specific provider using the provider/model syntax.

How to Call Models

Auto Routing (Python)
Auto Routing (TypeScript)
Force Provider (Python)
Force Provider (TypeScript)

Let the API choose the best provider automatically:

from lunar import Lunar

client = Lunar()

# API automatically selects the best provider
response = client.chat.completions.create(
    model="claude-3-7-sonnet",
    messages=[{"role": "user", "content": "Hello!"}]
)

Let the API choose the best provider automatically:

import { Lunar } from "lunar";

const client = new Lunar();

// API automatically selects the best provider
const response = await client.chat.completions.create({
  model: "claude-3-7-sonnet",
  messages: [{ role: "user", content: "Hello!" }],
});

Use provider/model syntax to route to a specific provider:

from lunar import Lunar

client = Lunar()

# Force Anthropic as the provider
response = client.chat.completions.create(
    model="anthropic/claude-3-7-sonnet",
    messages=[{"role": "user", "content": "Hello!"}]
)

# Force Bedrock as the provider
response = client.chat.completions.create(
    model="bedrock/claude-3-7-sonnet",
    messages=[{"role": "user", "content": "Hello!"}]
)

Use provider/model syntax to route to a specific provider:

import { Lunar } from "lunar";

const client = new Lunar();

// Force Anthropic as the provider
const response = await client.chat.completions.create({
  model: "anthropic/claude-3-7-sonnet",
  messages: [{ role: "user", content: "Hello!" }],
});

// Force Bedrock as the provider
const response2 = await client.chat.completions.create({
  model: "bedrock/claude-3-7-sonnet",
  messages: [{ role: "user", content: "Hello!" }],
});

Intelligent Routing

When you call a model without specifying a provider, the API automatically selects the best available provider based on real-time data:

Python
TypeScript

from lunar import Lunar

client = Lunar()

# No provider specified - API chooses the best one
response = client.chat.completions.create(
    model="llama-3.3-70b",
    messages=[{"role": "user", "content": "Hello!"}]
)

# The API might route to Groq, Bedrock, Cerebras, or SambaNova
# based on current performance metrics

import { Lunar } from "lunar";

const client = new Lunar();

// No provider specified - API chooses the best one
const response = await client.chat.completions.create({
  model: "llama-3.3-70b",
  messages: [{ role: "user", content: "Hello!" }],
});

// The API might route to Groq, Bedrock, Cerebras, or SambaNova
// based on current performance metrics

How Provider Selection Works

When multiple providers support the same model, the routing engine considers:

Factor	Description
Latency	Current response times from each provider
Availability	Provider health and uptime status
Cost	Token pricing differences between providers
Load	Current traffic and queue times

You only need to configure API keys for the providers you want to use. The routing engine will only consider providers where you have valid credentials configured.

Example: Llama Models

llama-3.3-70b is available on multiple providers. Without a prefix, the API routes intelligently:

Python
TypeScript

# Intelligent routing - best provider chosen automatically
response = client.chat.completions.create(
    model="llama-3.3-70b",
    messages=[{"role": "user", "content": "Explain quantum computing"}]
)

# If you need a specific provider:
response = client.chat.completions.create(
    model="groq/groq-llama-3.3-70b-versatile",  # Ultra-fast inference
    messages=[{"role": "user", "content": "Explain quantum computing"}]
)

// Intelligent routing - best provider chosen automatically
const response = await client.chat.completions.create({
  model: "llama-3.3-70b",
  messages: [{ role: "user", content: "Explain quantum computing" }],
});

// If you need a specific provider:
const response2 = await client.chat.completions.create({
  model: "groq/groq-llama-3.3-70b-versatile", // Ultra-fast inference
  messages: [{ role: "user", content: "Explain quantum computing" }],
});

Models by Provider

OpenAI Models

Use prefix: openai/

response = client.chat.completions.create(
    model="openai/gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)

GPT-4o Series

Model ID	Description
`gpt-4o`	Latest multimodal flagship model
`gpt-4o-mini`	Fast and cost-effective
`gpt-4o-2024-05-13`	Specific dated version
`gpt-4o-search-preview`	With web search capabilities
`gpt-4o-mini-search-preview`	Mini with web search

GPT-4.1 Series

Model ID	Description
`gpt-4.1`	Enhanced GPT-4
`gpt-4.1-mini`	Smaller, faster variant
`gpt-4.1-nano`	Fastest, most efficient

GPT-5 Series

Model ID	Description
`gpt-5`	Next generation flagship
`gpt-5-pro`	Professional tier
`gpt-5-mini`	Balanced performance
`gpt-5-nano`	Fast and efficient
`gpt-5-chat-latest`	Latest chat version
`gpt-5-search-api`	With search capabilities

GPT-5.1 Series

Model ID	Description
`gpt-5.1`	Latest GPT-5 update
`gpt-5.1-chat-latest`	Latest chat version
`gpt-5.1-codex`	Code specialized
`gpt-5.1-codex-mini`	Compact code model

o1 Reasoning Series

Model ID	Description
`o1`	Advanced reasoning model
`o1-mini`	Compact reasoning
`o1-pro`	Professional reasoning

o3 Reasoning Series

Model ID	Description
`o3`	Latest reasoning model
`o3-mini`	Compact variant
`o3-pro`	Professional tier
`o3-deep-research`	Deep research capabilities

o4 Series

Model ID	Description
`o4-mini`	Compact reasoning
`o4-mini-deep-research`	With deep research

Specialized Models

Model ID	Description
`codex-mini-latest`	Code generation
`computer-use-preview`	Computer use capabilities

Anthropic Models

Use prefix: anthropic/

response = client.chat.completions.create(
    model="anthropic/claude-sonnet-4-5",
    messages=[{"role": "user", "content": "Hello!"}]
)

Claude 4.5 Series (Latest)

Model ID	Description
`claude-opus-4-5`	Most capable, complex tasks
`claude-sonnet-4-5`	Balanced performance
`claude-haiku-4-5`	Fast and efficient

Claude 4 Series

Model ID	Description
`claude-opus-4`	High capability
`claude-opus-4-1`	Updated variant
`claude-sonnet-4`	Balanced model
`claude-sonnet-4-5`	Enhanced Sonnet

Claude 3.7 Series

Model ID	Description
`claude-3-7-sonnet`	Latest Sonnet (200K context)

Claude 3.5 Series

Model ID	Description
`claude-3-5-haiku`	Fast responses

Claude 3 Series

Model ID	Description
`claude-3-opus`	Most capable v3
`claude-3-haiku`	Fast and affordable

Google Gemini Models

Use prefix: gemini/

response = client.chat.completions.create(
    model="gemini/gemini-2.5-pro",
    messages=[{"role": "user", "content": "Hello!"}]
)

Gemini 3 Preview

Model ID	Description
`gemini-3-pro-preview`	Next gen professional
`gemini-3-flash-preview`	Next gen fast

Gemini 2.5 Series

Model ID	Description
`gemini-2.5-pro`	Professional tier (1M context)
`gemini-2.5-flash`	Fast responses
`gemini-2.5-flash-lite`	Lightweight version

Gemini 2.0 Series

Model ID	Description
`gemini-2.0-flash`	Fast multimodal
`gemini-2.0-flash-lite`	Lightweight version

Mistral Models

Use prefix: mistral/

response = client.chat.completions.create(
    model="mistral/mistral-large-latest",
    messages=[{"role": "user", "content": "Hello!"}]
)

Core Models

Model ID	Description
`mistral-large-latest`	Most capable
`mistral-medium-latest`	Balanced
`mistral-small-latest`	Fast and efficient

Magistral Series (Reasoning)

Model ID	Description
`magistral-medium-latest`	Reasoning medium
`magistral-small-latest`	Reasoning small

Code Models

Model ID	Description
`codestral-latest`	Code specialized

Open Models

Model ID	Description
`open-mistral-nemo`	Open source variant

DeepSeek Models

Use prefix: deepseek/

response = client.chat.completions.create(
    model="deepseek/deepseek-chat",
    messages=[{"role": "user", "content": "Hello!"}]
)

Model ID	Description
`deepseek-chat`	General chat model
`deepseek-reasoner`	Advanced reasoning

Perplexity Sonar Models

Use prefix: perplexity/

response = client.chat.completions.create(
    model="perplexity/sonar-pro",
    messages=[{"role": "user", "content": "What are the latest news?"}]
)

Model ID	Description
`sonar`	Base search model
`sonar-pro`	Enhanced search
`sonar-reasoning`	With reasoning
`sonar-reasoning-pro`	Pro reasoning
`sonar-deep-research`	In-depth research

Groq Models (Ultra-Fast Inference)

Use prefix: groq/

response = client.chat.completions.create(
    model="groq/groq-llama-3.3-70b-versatile",
    messages=[{"role": "user", "content": "Hello!"}]
)

Llama Models

Model ID	Description
`groq-llama-3.1-8b-instant`	Ultra fast 8B
`groq-llama-3.3-70b-versatile`	Versatile 70B
`groq-llama-4-scout`	Scout model
`groq-llama-4-maverick`	Maverick model

Qwen Models

Model ID	Description
`groq-qwen3-32b`	Qwen 32B

GPT-OSS Models

Model ID	Description
`groq-gpt-oss-20b`	Open source 20B
`groq-gpt-oss-120b`	Open source 120B

Other Models

Model ID	Description
`groq-kimi-k2-instruct`	Kimi K2 Instruct

Cerebras Models (Fast Inference)

Use prefix: cerebras/

response = client.chat.completions.create(
    model="cerebras/cerebras-llama-3.3-70b",
    messages=[{"role": "user", "content": "Hello!"}]
)

Llama Models

Model ID	Description
`cerebras-llama-3.1-8b`	Fast 8B
`cerebras-llama-3.3-70b`	Fast 70B

Qwen Models

Model ID	Description
`cerebras-qwen-3-32b`	Qwen 32B
`cerebras-qwen-3-235b`	Qwen 235B

Other Models

Model ID	Description
`cerebras-gpt-oss-120b`	GPT OSS 120B
`cerebras-zai-glm-4.6`	ZAI GLM 4.6

SambaNova Models

Use prefix: sambanova/

response = client.chat.completions.create(
    model="sambanova/sambanova-llama-3.3-70b",
    messages=[{"role": "user", "content": "Hello!"}]
)

Model ID	Description
`sambanova-llama-3.1-8b`	Llama 3.1 8B
`sambanova-llama-3.3-70b`	Llama 3.3 70B
`sambanova-qwen3-32b`	Qwen3 32B
`sambanova-gpt-oss-120b`	GPT OSS 120B
`sambanova-deepseek-r1`	DeepSeek R1
`sambanova-deepseek-v3`	DeepSeek V3

AWS Bedrock Models

Use prefix: bedrock/

response = client.chat.completions.create(
    model="bedrock/amazon-nova-pro",
    messages=[{"role": "user", "content": "Hello!"}]
)

Amazon Nova

Model ID	Description
`amazon-nova-micro`	Smallest, fastest
`amazon-nova-lite`	Lightweight
`amazon-nova-pro`	Professional
`amazon-nova-premier`	Most capable

Claude on Bedrock

Model ID	Description
`claude-3-5-haiku`	Claude 3.5 Haiku
`claude-3-7-sonnet`	Claude 3.7 Sonnet
`claude-3-haiku`	Claude 3 Haiku
`claude-3-opus`	Claude 3 Opus
`claude-3.5-haiku`	Claude 3.5 Haiku (alt)
`claude-3.5-sonnet`	Claude 3.5 Sonnet
`claude-3.7-sonnet`	Claude 3.7 Sonnet (alt)
`claude-haiku-4-5`	Claude Haiku 4.5
`claude-haiku-4.5`	Claude Haiku 4.5 (alt)
`claude-opus-4`	Claude Opus 4
`claude-opus-4.1`	Claude Opus 4.1
`claude-opus-4.5`	Claude Opus 4.5
`claude-sonnet-4`	Claude Sonnet 4
`claude-sonnet-4-5`	Claude Sonnet 4.5
`claude-sonnet-4.5`	Claude Sonnet 4.5 (alt)

Cohere on Bedrock

Model ID	Description
`command-r`	Command R base
`command-r+`	Enhanced Command R

DeepSeek on Bedrock

Model ID	Description
`deepseek-r1`	DeepSeek R1 reasoning

Gemma on Bedrock

Model ID	Description
`gemma-3-4b-it`	Gemma 3 4B Instruct
`gemma-3-12b-it`	Gemma 3 12B Instruct
`gemma-3-27b-pt`	Gemma 3 27B Pretrained

GPT-OSS on Bedrock

Model ID	Description
`gpt-oss-20b`	GPT OSS 20B
`gpt-oss-120b`	GPT OSS 120B

AI21 Jamba on Bedrock

Model ID	Description
`jamba-1.5-mini`	Jamba 1.5 Mini
`jamba-1.5-large`	Jamba 1.5 Large

Llama on Bedrock

Model ID	Description
`llama-3-1-8b`	Llama 3.1 8B
`llama-3-1-70b`	Llama 3.1 70B
`llama-3-1-405b`	Llama 3.1 405B
`llama-3-2-1b`	Llama 3.2 1B
`llama-3-2-3b`	Llama 3.2 3B
`llama-3-3-70b`	Llama 3.3 70B
`llama-3.1-8b-instruct`	Llama 3.1 8B Instruct
`llama-3.1-70b-instruct`	Llama 3.1 70B Instruct
`llama-3.2-1b-instruct`	Llama 3.2 1B Instruct
`llama-3.2-3b-instruct`	Llama 3.2 3B Instruct
`llama-3.2-11b-vision-instruct`	Llama 3.2 11B Vision
`llama-3.2-90b-vision-instruct`	Llama 3.2 90B Vision
`llama-3.3-70b-instruct`	Llama 3.3 70B Instruct
`llama-4-scout-17b`	Llama 4 Scout 17B
`llama-4-maverick-17b`	Llama 4 Maverick 17B

Mistral on Bedrock

Model ID	Description
`mistral-7b`	Mistral 7B
`mistral-small`	Mistral Small
`mistral-large`	Mistral Large
`mistral-large-3`	Mistral Large 3
`mixtral-8x7b`	Mixtral 8x7B MoE
`ministral-3-8b`	Ministral 3 8B
`ministral-14b-3.0`	Ministral 14B 3.0
`magistral-small-2509`	Magistral Small
`pixtral-large-25.02`	Pixtral Large (Vision)

Qwen on Bedrock

Model ID	Description
`qwen3-32b`	Qwen3 32B
`qwen3-coder-30b-a3b`	Qwen3 Coder 30B

Quick Reference

Provider Prefixes

Provider	Prefix	Example
OpenAI	`openai/`	`openai/gpt-4o`
Anthropic	`anthropic/`	`anthropic/claude-sonnet-4-5`
Google	`gemini/`	`gemini/gemini-2.5-pro`
Mistral	`mistral/`	`mistral/mistral-large-latest`
DeepSeek	`deepseek/`	`deepseek/deepseek-chat`
Perplexity	`perplexity/`	`perplexity/sonar-pro`
Groq	`groq/`	`groq/groq-llama-3.3-70b-versatile`
Cerebras	`cerebras/`	`cerebras/cerebras-llama-3.3-70b`
SambaNova	`sambanova/`	`sambanova/sambanova-llama-3.3-70b`
AWS Bedrock	`bedrock/`	`bedrock/amazon-nova-pro`

Models with Multiple Providers

Some models are available from multiple providers. The API automatically routes to the best available provider, or you can force a specific one:

Model	Available Providers
`claude-3-5-haiku`	anthropic, bedrock
`claude-3-7-sonnet`	anthropic, bedrock
`claude-3-haiku`	anthropic, bedrock
`claude-3-opus`	anthropic, bedrock
`claude-haiku-4-5`	anthropic, bedrock
`claude-opus-4`	anthropic, bedrock
`claude-sonnet-4`	anthropic, bedrock
`claude-sonnet-4-5`	anthropic, bedrock

Python
TypeScript

# Let API choose the best provider
response = client.chat.completions.create(
    model="claude-3-7-sonnet",
    messages=[{"role": "user", "content": "Hello!"}]
)

# Force Anthropic
response = client.chat.completions.create(
    model="anthropic/claude-3-7-sonnet",
    messages=[{"role": "user", "content": "Hello!"}]
)

# Force Bedrock
response = client.chat.completions.create(
    model="bedrock/claude-3-7-sonnet",
    messages=[{"role": "user", "content": "Hello!"}]
)

// Let API choose the best provider
const response = await client.chat.completions.create({
  model: "claude-3-7-sonnet",
  messages: [{ role: "user", content: "Hello!" }],
});

// Force Anthropic
const response2 = await client.chat.completions.create({
  model: "anthropic/claude-3-7-sonnet",
  messages: [{ role: "user", content: "Hello!" }],
});

// Force Bedrock
const response3 = await client.chat.completions.create({
  model: "bedrock/claude-3-7-sonnet",
  messages: [{ role: "user", content: "Hello!" }],
});

Model Selection Guidelines

Use Case	Recommended Models
Low cost	`gpt-4o-mini`, `claude-3-haiku`, `gemini-2.0-flash-lite`
Fast responses	`groq-llama-3.1-8b-instant`, `cerebras-llama-3.1-8b`
Complex reasoning	`o3`, `claude-opus-4-5`, `deepseek-reasoner`
Code generation	`gpt-5.1-codex`, `claude-sonnet-4-5`, `codestral-latest`
Large context	`gemini-2.5-pro` (1M), `claude-3-7-sonnet` (200K)
Web search	`sonar-pro`, `sonar-deep-research`, `gpt-4o-search-preview`
Research	`sonar-deep-research`, `o3-deep-research`

Getting Started

Lunar SDK

Pricing

​Supported Models

​How to Call Models

​Intelligent Routing

​How Provider Selection Works

​Example: Llama Models

​Models by Provider

​OpenAI Models

​GPT-4o Series

​GPT-4.1 Series

​GPT-5 Series

​GPT-5.1 Series

​o1 Reasoning Series

​o3 Reasoning Series

​o4 Series

​Specialized Models

​Anthropic Models

​Claude 4.5 Series (Latest)

​Claude 4 Series

​Claude 3.7 Series

​Claude 3.5 Series

​Claude 3 Series

​Google Gemini Models

​Gemini 3 Preview

​Gemini 2.5 Series

​Gemini 2.0 Series

​Mistral Models

​Core Models

​Magistral Series (Reasoning)

​Code Models

​Open Models

​DeepSeek Models

​Perplexity Sonar Models

​Groq Models (Ultra-Fast Inference)

​Llama Models

​Qwen Models

​GPT-OSS Models

​Other Models

​Cerebras Models (Fast Inference)

​Llama Models

​Qwen Models

​Other Models

​SambaNova Models

​AWS Bedrock Models

​Amazon Nova

​Claude on Bedrock

​Cohere on Bedrock

​DeepSeek on Bedrock

​Gemma on Bedrock

​GPT-OSS on Bedrock

​AI21 Jamba on Bedrock

​Llama on Bedrock

​Mistral on Bedrock

​Qwen on Bedrock

​Quick Reference

​Provider Prefixes

​Models with Multiple Providers

​Model Selection Guidelines

Supported Models

How to Call Models

Intelligent Routing

How Provider Selection Works

Example: Llama Models

Models by Provider

OpenAI Models

GPT-4o Series

GPT-4.1 Series

GPT-5 Series

GPT-5.1 Series

o1 Reasoning Series

o3 Reasoning Series

o4 Series

Specialized Models

Anthropic Models

Claude 4.5 Series (Latest)

Claude 4 Series

Claude 3.7 Series

Claude 3.5 Series

Claude 3 Series

Google Gemini Models

Gemini 3 Preview

Gemini 2.5 Series

Gemini 2.0 Series

Mistral Models

Core Models

Magistral Series (Reasoning)

Code Models

Open Models

DeepSeek Models

Perplexity Sonar Models

Groq Models (Ultra-Fast Inference)

Llama Models

Qwen Models

GPT-OSS Models

Other Models

Cerebras Models (Fast Inference)

Llama Models

Qwen Models

Other Models

SambaNova Models

AWS Bedrock Models

Amazon Nova

Claude on Bedrock

Cohere on Bedrock

DeepSeek on Bedrock

Gemma on Bedrock

GPT-OSS on Bedrock

AI21 Jamba on Bedrock

Llama on Bedrock

Mistral on Bedrock

Qwen on Bedrock

Quick Reference

Provider Prefixes

Models with Multiple Providers

Model Selection Guidelines