One standard API.
Every frontier & open model.
Switch from OpenAI to Anthropic, Google, or DeepSeek without refactoring your codebase. Simply point your existing OpenAI SDK to api100’s low-latency edge gateway.
// ~/.continue/config.json
// Install the "Continue" extension from VS Code marketplace, then add api100:
{
"models": [
{
"title": "api100 Claude 3.5 Sonnet",
"provider": "openai",
"model": "claude-3-5-sonnet",
"apiKey": "sk-api100-v1-YOUR_API_KEY",
"apiBase": "https://api.apihundred.com/v1"
},
{
"title": "api100 GPT-4o",
"provider": "openai",
"model": "gpt-4o",
"apiKey": "sk-api100-v1-YOUR_API_KEY",
"apiBase": "https://api.apihundred.com/v1"
},
{
"title": "api100 DeepSeek R1",
"provider": "openai",
"model": "deepseek-r1",
"apiKey": "sk-api100-v1-YOUR_API_KEY",
"apiBase": "https://api.apihundred.com/v1"
}
]
}Connect to VS Code, Cursor & Autonomous Agents
Use your single api100 key across your favorite development environments with full streaming and model flexibility.
VS Code — Continue
The #1 open-source AI extension for VS Code. Enjoy tab autocomplete, inline code refactoring (Ctrl+I), and sidebar chat.
config.json.apiBase to https://api.apihundred.com/v1 and add your key.VS Code — Cline / Roo
Autonomous coding agents that read codebases, create files, and run tests. Completely powered by api100.
https://api.apihundred.com/v1, Model: claude-3-5-sonnet.Cursor IDE
Connect Cursor directly to api100's gateway. Route codebase embeddings and multi-file reasoning to top models.
https://api.apihundred.com/v1.Unified API Endpoints
Deterministic routing specifications with full OpenAI JSON schema parity.
Generate streaming or buffered completions across all frontier models.
| Field | Type | Requirement | Description |
|---|---|---|---|
| model | string | required | Model slug (e.g., 'claude-3-5-sonnet', 'gpt-4o', 'deepseek-r1') |
| messages | array | required | Array of chat messages containing role ('system', 'user', 'assistant') and content |
| temperature | number | optional | Sampling temperature between 0.0 and 2.0 (default 0.7) |
| max_tokens | integer | optional | Maximum output tokens to generate before stopping |
| stream | boolean | optional | Whether to stream partial message deltas via Server-Sent Events (SSE) |
| tools | array | optional | List of tools the model may call for function execution |
| response_format | object | optional | Enforce JSON output schema: { type: 'json_object' } |
Retrieve real-time catalog of verified models, latency metrics, and pricing.
Inspect live token consumption ledger, wallet credits, and daily quota telemetry.
Error Handling & Status Codes
api100 returns standard RFC 7807 problem details with actionable error messages.
Request succeeded and completion stream initiated.
Malformed JSON payload or invalid model parameters.
Missing, expired, or malformed Bearer API key.
Insufficient compute credits in prepaid wallet.
Requests per minute (RPM) or token quota exceeded.
Upstream provider offline; automatically routed to fallback endpoint.
Tier Concurrency & Rate Limits
Guaranteed token throughput per plan. Upgrades apply immediately across all API keys.
| Plan Tier | Requests / Min (RPM) | Tokens / Min (TPM) | Concurrent Streams | Model Access |
|---|---|---|---|---|
| Free Starter | 20 RPM | 40,000 TPM | 2 concurrent | Tier 0 (Flash, Llama 3.3, 4o-mini) |
| api100 Pro | 120 RPM | 250,000 TPM | 10 concurrent | All frontier models (Claude 3.5, 4o, R1) |
| Team & Enterprise | 1,000+ RPM | 2,000,000+ TPM | 50+ concurrent | Dedicated inference queues + custom limits |
Ready to deploy frontier intelligence?
Generate an API key and begin routing completions in less than 60 seconds.