Documentation

The API is OpenAI-compatible: any OpenAI SDK, LangChain, LlamaIndex or plain HTTP works by changing two lines — the base URL and the key. Billing is prepaid, per token, in INR.

Quickstart

1. Create an account — trial credit included.
2. Copy your API key from the console.
3. Point your OpenAI client at https://api.bookmyhost.com/v1.

pip install openai

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_KEY",
    base_url="https://api.bookmyhost.com/v1",
)

r = client.chat.completions.create(
    model="llama-3.3-70b",
    messages=[{"role": "user", "content": "Namaste! What can you do?"}],
)
print(r.choices[0].message.content)

Authentication

Bearer token in the Authorization header. Keys are created and rotated in the console; you can hold several (per app / environment) and see spend per key.

Authorization: Bearer sk-…

Chat completions

curl https://api.bookmyhost.com/v1/chat/completions \
  -H "Authorization: Bearer $KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama-3.3-70b",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Write a haiku about the monsoon."}
    ],
    "max_tokens": 256,
    "temperature": 0.7
  }'

Available chat models: llama-3.3-70b, qwen3-coder-30b, qwen2.5-vl-7b-instruct, llama-3.1-8b, qwen3-8b — full list with live pricing on the models page.

Streaming

Standard SSE streaming. Add stream_options to receive token usage in the final chunk.

stream = client.chat.completions.create(
    model="llama-3.3-70b",
    messages=[{"role": "user", "content": "Tell me a story"}],
    stream=True,
    stream_options={"include_usage": True},
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

Tool / function calling

OpenAI-style tools with tool_choice: "auto" is supported on the chat models. Remember to send the tool result back as a role: "tool" message to complete the round-trip.

tools = [{
  "type": "function",
  "function": {
    "name": "get_weather",
    "parameters": {"type": "object", "properties": {"city": {"type": "string"}}}
  }
}]
r = client.chat.completions.create(model="llama-3.3-70b",
    messages=[{"role": "user", "content": "Weather in Mumbai?"}],
    tools=tools, tool_choice="auto")

Embeddings

r = client.embeddings.create(
    model="bge-m3",
    input=["Namaste world", "नमस्ते दुनिया"],
)
print(len(r.data[0].embedding))

Embedding requests bill input tokens only.

Reasoning models

Some models think before answering; the reasoning arrives separately in reasoning_content (or reasoning) and the answer in content. Reasoning consumes output tokens — set max_tokens generously or the answer may come back empty.

Billing & errors

• Prepaid INR credit; each request deducts (input tokens × input rate + output tokens × output rate) / 1M.
• 401 — bad or revoked key. 403 — model not in your plan tier or IP not allowed.
• 429 — over your requests-per-minute limit; back off and retry.
• Budget exceeded returns 400 with an ExceededBudget message — top up in the console.
• A request to a standby model waits a few seconds while it wakes; nothing fails.