Supported LLM Platforms

PipeLLM Gateway supports mainstream LLM service providers. Biggest advantage: Use official SDKs (Anthropic, OpenAI, Google, etc.), just change baseURL to access any platform transparently. No need to learn new APIs, zero code changes! Complete model list: https://www.pipellm.com/models

🤖 OpenAI Ecosystem

OpenAI Official

Status: ✅ Fully supported Usage: Use OpenAI official SDK

import openai
client = openai.OpenAI(
    api_key="your-pipellm-key",
    base_url="https://api.pipellm.com/v1"  # Point to our gateway
)

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

Supported models:

GPT-4 Series: gpt-4, gpt-4-turbo, gpt-4o, gpt-4o-mini
GPT-3.5 Series: gpt-3.5-turbo, gpt-3.5-turbo-16k
Embeddings: text-embedding-ada-002, text-embedding-3-small, text-embedding-3-large
Speech: whisper-1
Images: dall-e-3, dall-e-2

Azure OpenAI

Status: ✅ Fully supported Usage: Use OpenAI SDK, transparently call Azure services

import openai
client = openai.OpenAI(
    api_key="your-pipellm-key",
    base_url="https://api.pipellm.com/v1"  # Point to our gateway
)

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

Supported models:

GPT-4 Series: gpt-4, gpt-4-32k, gpt-4-turbo, gpt-4o
GPT-3.5 Series: gpt-35-turbo, gpt-35-turbo-16k
Embeddings: text-embedding-ada-002

🦙 Anthropic Claude

Status: ✅ Fully supported Usage: Use Anthropic official SDK

from anthropic import Anthropic
client = Anthropic(
    api_key="your-pipellm-key",
    base_url="https://api.pipellm.com/v1"  # Point to our gateway
)

response = client.messages.create(
    model="claude-3-sonnet",
    messages=[{"role": "user", "content": "Hello"}]
)

Supported models:

Claude 3 Series:
- claude-3-haiku - Fast, economical
- claude-3-sonnet - Balanced performance
- claude-3-opus - Highest quality
- claude-3-5-sonnet - Latest version

Claude features:

Long context processing (supports 200K tokens)
Excellent reasoning and analysis
Strong instruction following
Tool usage support

Transparent cross-platform calling:

Your SDK	Actual Platform	Notes
Anthropic SDK	AWS Bedrock	Auto converts to Bedrock protocol
Anthropic SDK	Google Vertex	Auto converts to Vertex protocol
Anthropic SDK	Azure	Auto converts to Azure protocol
Anthropic SDK	Anthropic Official	Direct call

🤖 Google Gemini

Status: ✅ Fully supported Usage: Use Google’s native libraries or standard API

import requests

headers = {
    "Authorization": f"Bearer your-pipellm-key",
    "Content-Type": "application/json"
}

data = {
    "model": "gemini-pro",
    "contents": [{"role": "user", "parts": [{"text": "Hello"}]}]
}

response = requests.post(
    "https://api.pipellm.com/v1/chat/completions",
    headers=headers,
    json=data
)

Supported models:

gemini-pro - General-purpose model
gemini-pro-vision - Multimodal model
gemini-ultra - Advanced model (if available)

Gemini features:

Multimodal capabilities (text + images)
Code generation optimization
Fast response
Google ecosystem integration

☁️ AWS Bedrock

Status: ✅ Fully supported Usage: Use native SDKs to call Bedrock services

# Use Anthropic SDK to call Claude on Bedrock
from anthropic import Anthropic

client = Anthropic(
    api_key="your-pipellm-key",
    base_url="https://api.pipellm.com/v1"  # Point to our gateway
)

response = client.messages.create(
    model="anthropic.claude-3-sonnet-20240229-v1:0",
    messages=[{"role": "user", "content": "Hello"}]
)

# Use OpenAI SDK to call Llama 3 on Bedrock
import openai
client = openai.OpenAI(
    api_key="your-pipellm-key",
    base_url="https://api.pipellm.com/v1"
)

response = client.chat.completions.create(
    model="meta.llama3-70b-instruct-v1:0",
    messages=[{"role": "user", "content": "Write a React component"}]
)

Supported models:

Anthropic: claude-3-haiku, claude-3-sonnet, claude-3-opus
Amazon Titan: amazon.titan-text-express-v1, amazon.titan-text-lite-v1
AI21 Labs: ai21.j2-mid, ai21.j2-ultra
Cohere: cohere.command-text-v14, cohere.command-light-text-v14
Meta: meta.llama3-8b-instruct, meta.llama3-70b-instruct, meta.llama3-1-8b-instruct, meta.llama3-1-70b-instruct, meta.llama3-2-11b-vision, meta.llama3-2-90b-vision
Mistral: mistral.mistral-7b-instruct, mistral.mixtral-8x7b-instruct, mistral.mistral-large-latest, mistral.mistral-small-latest
Stability AI: stability.stable-diffusion-xl-v1

Bedrock advantages:

AWS native integration
Enterprise-level security
Scalability
Pay-per-use

🌐 Other Cloud Platforms

Google Vertex AI

Status: ✅ Fully supported

Fireworks AI

Status: ✅ Fully supported Supported models:

accounts/fireworks/models/firefunction-v2
accounts/fireworks/models/llama-v3p1-405b
accounts/fireworks/models/llama-v3p1-70b
accounts/fireworks/models/llama-v3p1-8b
accounts/stabilityai/models/stable-diffusion-xl-1024-v1-0
Other Fireworks models

Together AI

Status: ✅ Fully supported Supported models:

Nous-Hermes-2-Mixtral-8x7B-DPO
Llama-3-8b-SFT
CodeLlama-34b
WizardLM-2-8x22b
Other Together AI models

Groq

Status: ✅ Fully supported Supported models:

llama3-8b-8192
llama3-70b-8192
mixtral-8x7b-32768
gemma-7b-it
gemma2-9b-it
llama-3-3-70b-versatile
llama-3-3-8b-instant

Advantages:

Ultra-fast inference
Low latency
Real-time application optimization

Replicate

Status: ✅ Fully supported Supported features:

Image generation
Video generation
Audio processing
Custom model deployment

OpenRouter

Status: ✅ Fully supported Features:

Aggregate multiple providers
Unified billing
Simplified access

🎨 Media Processing Platforms

Stability AI

Status: ✅ Fully supported Supported models:

stable-diffusion-xl-1024-v1-0
stable-diffusion-3
stable-cascade
stable-video-diffusion

Ideogram

Status: ✅ Fully supported Features:

Creative image generation
Text rendering optimization
Artistic styles

Luma Labs

Status: ✅ Fully supported Supported features:

3D model generation
Image-to-3D conversion
Video processing

📊 Platform Comparison

Feature	OpenAI	Anthropic	Gemini	AWS Bedrock	Azure
Max Context	128K	200K	32K	200K	128K
Multimodal	✅	❌	✅	Partial	Partial
Code Ability	⭐⭐⭐⭐⭐	⭐⭐⭐⭐	⭐⭐⭐⭐	⭐⭐⭐	⭐⭐⭐⭐
Reasoning	⭐⭐⭐⭐	⭐⭐⭐⭐⭐	⭐⭐⭐⭐	⭐⭐⭐⭐	⭐⭐⭐⭐
Speed	Fast	Very Fast	Very Fast	Fast	Fast
Price	High	Medium	Low	Medium	Medium
Enterprise	✅	✅	✅	✅	✅

🚀 How to Choose the Right Platform

1. By Use Case

Code Generation:

Best: OpenAI GPT-4o, Claude 3
Features: High accuracy, multi-language support

Long Document Processing:

Best: Claude 3 (200K context)
Features: Can process entire books or codebases

Creative Writing:

Best: OpenAI GPT-4o, Gemini Pro
Features: High creativity, diverse styles

Enterprise Applications:

Best: Azure OpenAI, AWS Bedrock
Features: Enterprise security, data guarantees

Cost-Sensitive:

Best: Gemini Pro, Llama 3
Features: High cost-effectiveness

2. By Technical Requirements

Multimodal Needs:

OpenAI GPT-4o
Google Gemini Pro Vision
AWS Titan Multimodal

Long Context:

Anthropic Claude 3 (200K)
OpenAI GPT-4o (128K)
AWS Claude 3 (200K)

Fast Response:

Groq (hardware acceleration)
OpenAI (optimized network)
Anthropic (fast models)

3. Using PipeLLM Gateway Advantages

No Manual Selection:

# Gateway auto-selects
curl https://api.pipellm.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{"model": "auto", "messages": [...]}'

Smart Routing:

Auto-select based on load
Route based on model availability
Optimize based on cost

Failover:

Auto-switch if provider unavailable
Ensure service continuity
Reduce downtime risk

🛠️ Advanced Configuration

1. Specify Provider Preference

curl https://api.pipellm.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "X-Preferred-Provider: openai" \
  -d '{"model": "auto", "messages": [...]}'

2. Force Specific Provider

curl https://api.pipellm.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "X-Force-Provider: anthropic" \
  -d '{"model": "auto", "messages": [...]}'

3. Model Aliases

{
  "model": "best",        // Best quality
  "model": "fast",        // Fastest
  "model": "cheap",       // Most economical
  "model": "balanced"     // Balanced performance
}

4. Region Selection

curl https://api.pipellm.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "X-Region: us-east-1" \
  -d '{"model": "auto", "messages": [...]}'

📈 Performance Monitoring

Monitor via management dashboard:

Success rate comparison across providers
Average response time
Cost analysis
Model usage statistics

🤝 Support

If you need platform support or encounter issues:

Check documentation: Visit provider’s official docs
Enable debug mode: Use X-Debug: true to view details
Contact support: Email [email protected]

Tip: PipeLLM Gateway continuously adds support for new platforms and models. Stay updated!

Quick Start

Gateway Features

Supported LLM Platforms

Supported LLM Platforms

🤖 OpenAI Ecosystem

OpenAI Official

Azure OpenAI

🦙 Anthropic Claude

🤖 Google Gemini

☁️ AWS Bedrock

🌐 Other Cloud Platforms

Google Vertex AI

Fireworks AI

Together AI

Groq

Replicate

OpenRouter

🎨 Media Processing Platforms

Stability AI

Ideogram

Luma Labs

📊 Platform Comparison

🚀 How to Choose the Right Platform

1. By Use Case

2. By Technical Requirements

3. Using PipeLLM Gateway Advantages

🛠️ Advanced Configuration

1. Specify Provider Preference

2. Force Specific Provider

3. Model Aliases

4. Region Selection

📈 Performance Monitoring

🤝 Support

Quick Start

Gateway Features

​Supported LLM Platforms

​🤖 OpenAI Ecosystem

​OpenAI Official

​Azure OpenAI

​🦙 Anthropic Claude

​🤖 Google Gemini

​☁️ AWS Bedrock

​🌐 Other Cloud Platforms

​Google Vertex AI

​Fireworks AI

​Together AI

​Groq

​Replicate

​OpenRouter

​🎨 Media Processing Platforms

​Stability AI

​Ideogram

​Luma Labs

​📊 Platform Comparison

​🚀 How to Choose the Right Platform

​1. By Use Case

​2. By Technical Requirements

​3. Using PipeLLM Gateway Advantages

​🛠️ Advanced Configuration

​1. Specify Provider Preference

​2. Force Specific Provider

​3. Model Aliases

​4. Region Selection

​📈 Performance Monitoring

​🤝 Support

Supported LLM Platforms

🤖 OpenAI Ecosystem

OpenAI Official

Azure OpenAI

🦙 Anthropic Claude

🤖 Google Gemini

☁️ AWS Bedrock

🌐 Other Cloud Platforms

Google Vertex AI

Fireworks AI

Together AI

Groq

Replicate

OpenRouter

🎨 Media Processing Platforms

Stability AI

Ideogram

Luma Labs

📊 Platform Comparison

🚀 How to Choose the Right Platform

1. By Use Case

2. By Technical Requirements

3. Using PipeLLM Gateway Advantages

🛠️ Advanced Configuration

1. Specify Provider Preference

2. Force Specific Provider

3. Model Aliases

4. Region Selection

📈 Performance Monitoring

🤝 Support