Deploy AI agents.
Keep it cheap.
A simple API to run, orchestrate, and persist AI agents. Supports every major model. Start free, scale when you need to.
Free tier. No credit card required.
# submit a run
curl -X POST https://aaasp.ai/v1/runs \
-H "Authorization: Bearer $AAASP_API_KEY" \
-H "Content-Type: application/json" \
-d '{"deployment_id":"dep_abc","prompt":"Summarize this doc"}'
# response
{"id":"run_xyz","status":"queued","created_at":"2026-03-12T..."}How it works
Register & get a key
Sign up with your email. Your API key is issued instantly — no setup required.
Define your agent
Choose a model, write a system prompt, pick an executor. Deploy in one API call.
Run & observe
Submit prompts, stream results, receive webhooks. Monitor everything from the dashboard.
Everything you need
Async runs
Submit and poll, or use sync mode for immediate results. Webhook delivery included.
Session memory
Multi-turn conversations with automatic context windowing and persistent checkpointing.
Multi-model
Claude, GPT, Mistral, Groq, Together — swap models without changing your integration.
Batch processing
Submit hundreds of prompts atomically. All-or-nothing transactions, tracked per-run.
NPC Studio
Generate richly-characterised game NPCs. Smart (LLM) or dumb (pre-compiled, zero runtime cost).
Credit billing
Pay-as-you-go credits. LLM costs passed through at actual provider rates plus a small margin.
Simple pricing
LLM costs billed at provider rate + 3%. Nothing hidden.
Builder
- 5,000 runs / month
- All executors
- LLM at cost + 3%
- Streaming + webhooks
Scale
- 25,000 runs / month
- Priority queue
- LLM at cost + 3%
- 200 concurrent runs
Enterprise / custom pricing — get in touch