This gateway provides 100% official Anthropic Messages API emulation (/v1/messages, /v1/models, /v1/messages/count_tokens) with real-time SSE streaming, automatic persona spoofing, and intelligent token optimization.
https://claude-gateway.sarimtools.com/v1
Configure Anthropic's official Claude Code CLI by setting the environment variables in your terminal or shell profile (~/.zshrc or ~/.bashrc):
# Configure Anthropic Enterprise Gateway Endpoint
export ANTHROPIC_BASE_URL="https://claude-gateway.sarimtools.com/v1"
export ANTHROPIC_API_KEY="sk-ant-api03-YOUR_GATEWAY_KEY"
# Launch Claude Code CLI
claude
# Or launch with a specific enterprise model
claude --model claude-fable-5-2
• Compatible with Claude Code's model discovery, token count status bar (/v1/messages/count_tokens), and multi-turn tool calling.
You can route your native Claude Desktop Application through this enterprise gateway using third-party inference configuration.
Add or edit your Claude Desktop configuration file:
• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"api_endpoint": "https://claude-gateway.sarimtools.com/v1",
"api_key": "sk-ant-api03-YOUR_GATEWAY_KEY",
"default_model": "claude-fable-5-2"
}
Use the official anthropic Python library with zero code refactoring by specifying base_url:
import anthropic
client = anthropic.Anthropic(
base_url="https://claude-gateway.sarimtools.com/v1",
api_key="sk-ant-api03-YOUR_GATEWAY_KEY"
)
# 1. Standard Synchronous Call
response = client.messages.create(
model="claude-fable-5-2",
max_tokens=1024,
messages=[
{"role": "user", "content": "Explain quantum computing in two sentences."}
]
)
print(response.content[0].text)
# 2. Real-Time Streaming Call
with client.messages.stream(
model="claude-sonnet-5-5",
max_tokens=1024,
messages=[{"role": "user", "content": "Write a Python script to sort a dictionary."}]
) as stream:
for text in stream.text_stream:
print(text, end="", flush=True)
Execute HTTP requests directly via curl with streaming enabled:
curl -N "https://claude-gateway.sarimtools.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: sk-ant-api03-YOUR_GATEWAY_KEY" \
-H "anthropic-version: 2023-06-01" \
-d '{
"model": "claude-fable-5-2",
"max_tokens": 1024,
"stream": true,
"messages": [
{"role": "user", "content": "Hello Claude! What model are you?"}
]
}'
Compatible with @anthropic-ai/sdk:
import Anthropic from '@anthropic-ai/sdk';
const anthropic = new Anthropic({
baseURL: 'https://claude-gateway.sarimtools.com/v1',
apiKey: 'sk-ant-api03-YOUR_GATEWAY_KEY',
});
async function main() {
const message = await anthropic.messages.create({
model: 'claude-fable-5-2',
max_tokens: 1024,
messages: [{ role: 'user', content: 'Say hello!' }],
});
console.log(message.content[0]);
}
main();
X-Accel-Buffering: no and explicit buffer flushing for smooth, sub-millisecond real-time typing experiences.