One API key.
Every model.

The Yielding Bear API provides unified access to 100+ large language models through a single OpenAI-compatible endpoint. Just change your base URL — no code refactoring required.

Our intelligent routing automatically selects the cheapest and fastest model for your reasoning needs — delivering 60-80% cost savings vs direct API calls. Every request is analyzed, classified, and routed to the optimal model.

The API supports both direct model selection (you pick the exact model) and automatic routing (YB picks the best model for your task). Both modes are available through the same endpoint.

All API requests require an API key passed as a Bearer token in the Authorization header. Get your API key from the dashboard.

Authorization: Bearer yb_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

export YB_API_KEY="yb_live_your_key_here"

# Verify your key works
curl https://api.yieldingbear.com/v1/models   -H "Authorization: Bearer $YB_API_KEY"

The Chat Completions endpoint is fully OpenAI-compatible. Send a POST request with messages and receive a structured response. Works with all OpenAI SDKs.

POST /v1/chat/completions

{
  "model": "groq/llama-3.1-8b-instant",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": "What is the capital of France?" }
  ],
  "temperature": 0.7,
  "max_tokens": 256,
  "stream": false
}

curl -X POST https://api.yieldingbear.com/v1/chat/completions   -H "Authorization: Bearer $YB_API_KEY"   -H "Content-Type: application/json"   -d '{
    "model": "groq/llama-3.1-8b-instant",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

{
  "id": "chatcmpl_xxxxxxxxxxxx",
  "object": "chat.completion",
  "created": 1717000000,
  "model": "groq/llama-3.1-8b-instant",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "The capital of France is Paris."
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 15,
    "completion_tokens": 12,
    "total_tokens": 27
  }
}

Generate text embeddings using any supported model. Fully compatible with OpenAI's embeddings API. Embeddings are useful for semantic search, similarity matching, and RAG (Retrieval Augmented Generation) pipelines.

POST /v1/embeddings

{
  "model": "Cohere/embed-english-v3.0",
  "input": "The food was delicious and the service was excellent."
}

curl -X POST https://api.yieldingbear.com/v1/embeddings   -H "Authorization: Bearer $YB_API_KEY"   -H "Content-Type: application/json"   -d '{
    "model": "Cohere/embed-english-v3.0",
    "input": "The food was delicious and the service was excellent."
  }'

List all available models. This endpoint is public and does not require authentication. Use it to discover models and check their current pricing.

GET /v1/models

curl https://api.yieldingbear.com/v1/models

Rate limits protect the API from abuse and ensure fair access. Limits are applied per API key and vary by tier.

Rate limit headers are included in every response:
X-RateLimit-Limit: 60 X-RateLimit-Remaining: 59 X-RateLimit-Reset: 1717000060

Enable streaming to receive tokens as they are generated. This reduces perceived latency for long responses. Use the stream: true parameter.

{
  "model": "groq/llama-3.1-8b-instant",
  "messages": [{"role": "user", "content": "Write a haiku about coding"}],
  "stream": true
}

data: {"id":"1","choices":[{"delta":{"content":"While"},"index":0}]}

data: {"id":"1","choices":[{"delta":{"content":" my"},"index":0}]}

data: {"id":"1","choices":[{"delta":{"content":" fingers"},"index":0}]}

data: {"id":"1","choices":[{"finish_reason":"stop","index":0,"delta":{}}]}

data: [DONE]

from openai import OpenAI

client = OpenAI(
    api_key=os.environ["YB_API_KEY"],
    base_url="https://api.yieldingbear.com/v1",
)

stream = client.chat.completions.create(
    model="groq/llama-3.1-8b-instant",
    messages=[{"role": "user", "content": "Hello!"}],
    stream=True,
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

Process multiple prompts in a single API call. Batch requests are more efficient than making individual calls and can reduce overall latency for bulk operations.

POST /v1/batch

curl -X POST https://api.yieldingbear.com/v1/batch   -H "Authorization: Bearer $YB_API_KEY"   -H "Content-Type: application/json"   -d '{
    "requests": [
      {"model": "groq/llama-3.1-8b-instant", "messages": [{"role": "user", "content": "Hi"}]},
      {"model": "groq/llama-3.1-8b-instant", "messages": [{"role": "user", "content": "Bye"}]}
    ]
  }'

Built-in safety and content filtering on every request. Guardrails run at the API layer — they add zero latency and cost nothing extra.

{
  "error": {
    "message": "Request blocked by guardrails: prompt injection detected",
    "type": "content_policy_violation",
    "code": "prompt_injection_blocked"
  }
}

The API follows OpenAI's error format. All errors return a JSON body with a top-level error object. Always check for errors in your response handling.

{
  "error": {
    "message": "Invalid API key",
    "type": "invalid_request_error",
    "code": "authentication_error",
    "param": null,
    "status": 401
  }
}

Free tier includes unlimited requests to Groq, Cerebras, and Together models. Paid tier gives access to premium models (OpenAI, Anthropic, Google, DeepSeek, Mistral) with a 20% markup over provider pricing.

Use any OpenAI-compatible SDK. Just change the base URL and add your API key. Official SDKs and community libraries are all supported.

Python

OpenAI SDKLangChainLlamaIndexDSPy

JavaScript

OpenAI SDK (Node)Vercel AI SDKLangChain.js

Go

go-openaiLangChain Go

Rust

rust-openaireqwest + serde

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="groq/llama-3.1-8b-instant",
    api_key=os.environ["YB_API_KEY"],
    base_url="https://api.yieldingbear.com/v1",
)

response = llm.invoke("Hello!")

Frequently Asked Questions