Coming Soon — The Haders Developer API is in active development. Join our Discord to get notified at launch.
OpenAI-compatible. Streaming. Per-token billing. Models that don't refuse your professional queries.
4
models
65 536
context tokens
$0.80
per 1M input tokens (Charon)
SSE
streaming on all models
Features
OpenAI-compatible
Drop-in replacement endpoint. Point your existing OpenAI SDK calls at api.haders.site and change the model name — nothing else.
Streaming SSE
Server-sent events streaming on all models. Tokens arrive at your client as they are generated — no polling, no waiting.
Per-token billing
Pay only for tokens consumed. No seat fees, no monthly minimums. Input and output priced separately per million tokens.
No content lobotomy
Haders models engage with the full problem space. No reflexive refusals on legitimate security, research, or professional queries.
System prompt support
Full system prompt control. Set personas, inject context, define output formats — the model follows your instructions.
Usage dashboard
Track token consumption, spending, and per-key breakdowns in real time from your developer dashboard.
Pricing
Prices per 1 million tokens, billed monthly in arrears.
| Model | Description | Input / 1M | Output / 1M | Best for |
|---|---|---|---|---|
| Charon | Fast, sharp, high-volume | $0.80 | $4.00 | Recon, lookups, rapid Q&A |
| Hermes | Balanced reasoning | $3.00 | $15.00 | Analysis, documentation, structured output |
| Cerberus | Extended autonomous reasoning | $15.00 | $75.00 | Long-horizon tasks, agentic workflows |
Get started
Sign in to your Haders account to generate an API key. Early access keys are available now — full developer billing launches soon.