Integrations

Works with every provider,
framework, and model.

One endpoint. One API key. Swap base_url to api.explane.ai/v1 and your existing code instantly routes across 52 AI providers.

api.explane.ai/v1 52 providers connected
Provider
Models
Req / hr
Avg latency
Status
OpenAI
gpt-4o, gpt-4o-mini, o1 +4
2,418
28ms
LIVE
Anthropic
claude-3.5-sonnet, haiku +2
1,832
34ms
LIVE
Google Gemini
gemini-2.0-flash, pro +3
948
41ms
LIVE
Mistral AI
mistral-large, codestral +3
312
52ms
LIVE
AWS Bedrock
titan, llama-3, claude +6
188
78ms
LIVE
DeepSeek
deepseek-r1, deepseek-v3 +1
520
22ms
LIVE
4,218 req / min
52 providers
99.97% uptime
0 errors
Supported Providers

52 AI providers.
Unified under one endpoint.

Unified endpoint api.explane.ai/v1
LLM Providers
OpenAI Anthropic Google Gemini Mistral AI Meta Llama Cohere DeepSeek Perplexity xAI Grok AI21 Amazon Titan Nous Hermes
Cloud AI Platforms
AWS Bedrock Azure OpenAI Google Vertex AI IBM watsonx
Embedding Models
OpenAI Embeddings Cohere Embed Voyage AI Jina AI Nomic Embed
Image & Multimodal
DALL·E 3 Stable Diffusion Ideogram Flux LLaVA
SDK & Setup

OpenAI-compatible.
30-second setup.

Explane speaks the OpenAI API. Change one line in your existing code — just swap base_url — and you're routing across all providers immediately.

  • Native SDKs: Python, Node.js, Go, Ruby, Java
  • Drop-in compatible with the official OpenAI SDK
  • Works with LangChain, LlamaIndex, CrewAI, and more
Already using OpenAI SDK?
You're 30 seconds away. Change base_url to https://api.explane.ai/v1 and keep everything else exactly as-is.
# pip install explane from explane import Explane ai = Explane(api_key="ex_...") response = ai.chat.completions.create( model="auto", messages=[{"role": "user", "content": "Hello"}], routing={"optimize": "cost"} )
— or use as an OpenAI drop-in (zero other changes) —
from openai import OpenAI client = OpenAI( api_key="ex_...", base_url="https://api.explane.ai/v1" # ← only change )
// npm install @explane/sdk import { Explane } from '@explane/sdk' const ai = new Explane({ apiKey: 'ex_...' }) const res = await ai.chat.completions.create({ model: 'auto', messages: [{ role: 'user', content: 'Hello' }] })
— or use as an OpenAI drop-in —
import OpenAI from 'openai' const client = new OpenAI({ apiKey: 'ex_...', baseURL: 'https://api.explane.ai/v1' // ← only change })
// go get github.com/explane/explane-go import ( explane "github.com/explane/explane-go" "context"; "os" ) client := explane.New(os.Getenv("EXPLANE_API_KEY")) resp, err := client.Chat.Completions.Create( context.Background(), explane.ChatCompletionRequest{ Model: "auto", Messages: []explane.Message{ {Role: "user", Content: "Hello"}, }, }, )
# Chat completion curl https://api.explane.ai/v1/chat/completions \ -H "Authorization: Bearer ex_..." \ -H "Content-Type: application/json" \ -d '{ "model": "auto", "messages": [{"role": "user", "content": "Hello"}], "x-explane-routing": {"optimize": "cost"} }'
Framework Integrations

Works with your existing AI framework.

Explane is OpenAI API-compatible — it drops into every major AI framework with zero code changes. Just point base_url at Explane.

LangChain
Route LangChain chains through Explane for automatic provider selection and cost optimization.
export OPENAI_BASE_URL=https://api.explane.ai/v1 export OPENAI_API_KEY=ex_...
OpenAI-compatible
LlamaIndex
Use Explane as your OpenAI-compatible LLM provider in any LlamaIndex pipeline or RAG system.
from llama_index.llms.openai import OpenAI llm = OpenAI( api_base="https://api.explane.ai/v1" )
OpenAI-compatible
CrewAI
Drop in Explane as the base LLM for all agents and tasks in your CrewAI workflow.
from crewai import Agent agent = Agent( llm="explane/auto", role="Researcher" )
Native integration
AutoGen
Compatible with AutoGen's OpenAI config — point base_url to Explane and all agents route through it.
config = [{ "model": "auto", "api_key": "ex_...", "base_url": "https://api.explane.ai/v1" }]
OpenAI-compatible
Semantic Kernel
Point the kernel's OpenAI connector to api.explane.ai for intelligent model routing across calls.
kernel.add_chat_service( "explane", OpenAIChatCompletion( endpoint="https://api.explane.ai/v1" ) )
OpenAI-compatible
Haystack
Use Explane as an OpenAI-compatible backend for any Haystack pipeline or RAG workflow.
generator = OpenAIGenerator( api_base_url="https://api.explane.ai/v1", api_key=Secret.from_env_var("EXPLANE_KEY") )
OpenAI-compatible
Observability

Native OpenTelemetry. Send traces
to your existing stack.

Explane exports OTLP traces automatically. Every LLM request becomes a traceable span with model, tokens, cost, routing decision, and provider info attached.

Datadog Grafana Tempo New Relic Honeycomb Jaeger Zipkin AWS X-Ray Dynatrace Elastic APM
Span attributes on every request
llm.model"gpt-4o"
llm.provider"openai"
llm.tokens.total1024
llm.cost.usd0.0031
explane.routing.reason"cost"
explane.latency.ms187
# Configure OTLP export export OTEL_EXPORTER_OTLP_ENDPOINT=https://otel.your-backend.com export OTEL_SERVICE_NAME=my-ai-app export EXPLANE_TRACES=true # Every request is traced automatically. # No instrumentation code required. # Spans include model, tokens, cost, # routing decision, and provider fallbacks.

Your stack. Our routing.

One line of code to connect 52 providers.