Quickstart
Moving to apicuts is one base URL change, and rolling back is the same line. Sign up at /signup to generate a company key in /app/keys, or book a free 360° cost report if you want a usage audit first.
Before you start
- Create an account at /signup, top up your prepaid workspace wallet in
/app/funding, and generate a company inference key (ack_...) in/app/keys. - Pick the base URL for your client protocol:
https://apicuts.ai/v1for OpenAI Responses, Chat Completions, and OpenAI-compatible Gemini routes; orhttps://apicuts.aifor Anthropic SDKs and Claude Code (which append/v1/messagesautomatically). - Optional: install the macOS Companion V1 for automatic Claude Desktop and Claude Code failover, or book a free 360° cost report to review your existing usage export.
1. Store your endpoint and key
export APICUTS_BASE_URL="https://apicuts.ai/v1" # OpenAI Responses & Chat Completions
export ANTHROPIC_BASE_URL="https://apicuts.ai" # Anthropic SDKs & Claude Code
export APICUTS_API_KEY="ack_..." # Company key from /app/keys2. Fetch live model IDs
Query GET /v1/models with your company key to see the literal model IDs and supported protocols enabled for your workspace (for example gpt-6.1-sol or claude-sonnet-5-5). Pass anthropic-version: 2023-06-01 to list Anthropic Messages models in Anthropic's pagination format.
curl https://apicuts.ai/v1/models \
-H "Authorization: Bearer $APICUTS_API_KEY"
# Anthropic Messages catalog format
curl https://apicuts.ai/v1/models \
-H "x-api-key: $APICUTS_API_KEY" \
-H "anthropic-version: 2023-06-01"3. Point your client at apicuts
apicuts serves native OpenAI Responses (/v1/responses), OpenAI Chat Completions (/v1/chat/completions), and native Anthropic Messages (/v1/messages) without translating protocols. All examples stream over SSE; Chat Completions also requires a fresh Idempotency-Key header per generation.
import os
from openai import OpenAI
client = OpenAI(
base_url=os.environ["APICUTS_BASE_URL"], # https://apicuts.ai/v1
api_key=os.environ["APICUTS_API_KEY"],
)
stream = client.responses.create(
model="gpt-6.1-sol",
input="Summarise yesterday's deploy log.",
stream=True,
)
for event in stream:
if event.type == "response.output_text.delta":
print(event.delta, end="", flush=True)Responses requires stream=True and defaults to store=false (responses are not stored server-side). Send full client-owned history for multi-turn calls.
import os
import anthropic
client = anthropic.Anthropic(
base_url=os.environ["ANTHROPIC_BASE_URL"], # https://apicuts.ai (do not add /v1)
api_key=os.environ["APICUTS_API_KEY"],
)
with client.messages.stream(
model="claude-sonnet-5-5",
max_tokens=1024,
system="Be concise.",
messages=[{"role": "user", "content": "Say hello"}],
) as stream:
for text in stream.text_stream:
print(text, end="", flush=True)Anthropic SDKs append /v1/messages and send anthropic-version: 2023-06-01 automatically. Always put system instructions in the top-level system field, not as a mid-conversation role: "system" message.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: process.env.APICUTS_BASE_URL, // https://apicuts.ai/v1
apiKey: process.env.APICUTS_API_KEY,
});
const stream = await client.chat.completions.create(
{
model: "gpt-6.1-sol",
messages: [{ role: "user", content: "Say hello" }],
max_tokens: 256,
stream: true,
},
{ headers: { "Idempotency-Key": crypto.randomUUID() } },
);
for await (const chunk of stream) {
process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}# Chat Completions (stream: true + Idempotency-Key required)
curl -N "https://apicuts.ai/v1/chat/completions" \
-H "Authorization: Bearer $APICUTS_API_KEY" \
-H "Idempotency-Key: $(uuidgen)" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-6.1-sol", "messages": [{"role": "user", "content": "Say hello"}], "max_tokens": 256, "stream": true}'
# Responses (stream: true required)
curl -N "https://apicuts.ai/v1/responses" \
-H "Authorization: Bearer $APICUTS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-6.1-sol", "input": "Say hello", "stream": true}'
# Anthropic Messages (x-api-key or Authorization: Bearer, plus anthropic-version)
curl -N "https://apicuts.ai/v1/messages" \
-H "x-api-key: $APICUTS_API_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "Content-Type: application/json" \
-d '{"model": "claude-sonnet-5-5", "max_tokens": 1024, "system": "Be concise.", "messages": [{"role": "user", "content": "Say hello"}], "stream": true}'Or install the macOS Companion V1
For Claude Desktop and Claude Code on macOS, install the menu-bar Auto-Hybrid router daemon. It uses your existing Pro/Max subscription first and fails over to apicuts automatically when quota hits 100%:
curl -fsSL https://apicuts.ai/install | sh -s -- --key "$APICUTS_API_KEY"4. Check the first requests
Responses come back in the exact native protocol format your SDK expects. Within seconds, requests appear in /app/requests and your savings dashboard with model, provider, latency, tokens and cost.
Roll back anytime
Set the base URL back to your provider's. Traffic skips apicuts entirely. No migration, no lock-in, no long-term contract.
Next
- Use apicuts with LangChain, LlamaIndex or the Vercel AI SDK
- See how requests are routed
- Set budgets, limits and alerts
- Estimate Claude Code session & monthly team API costs
- Read the Claude Code team cost guide
Want us to find your 20%+?
Get a free 360° cost report. Bring last month's provider usage and we'll show you your biggest waste and the first fix we'd ship.
Get a Free 360° Cost Report