Three flexible plans, from free browsing to professional subscription.
30 science topic sites with structured scientific knowledge data, updated daily.
slb_-prefixed key| Feature | Free | Partner | Pro |
|---|---|---|---|
| Rate limit | 60/min | 120/min | 600/min |
| Domain metadata | ✓ | ✓ | ✓ |
| Entity metadata | ✓ | ✓ | ✓ |
| Full-field entity data | ✕ | ✓ | ✓ |
| Open-source project registry | ✓ | ✓ | ✓ |
Semantic search /v1/search/semantic | ✕ | Partial | ✓ |
| Citation export | ✕ | ✕ | ✓ |
| Dedicated gateway | ✕ | ✓ (slb_) | ✓ (gtk_) |
| Data update notifications | ✕ | ✕ | ✓ |
| Technical support | Community | Dedicated channel | Priority response |
| Price | Free | Custom | ¥39.9/yr · ¥199 lifetime |
slb_ key)
120 requests / minute
gtk_ key)
600 requests / minute
Rate limits use a sliding window. When exceeded, the API returns HTTP 429 with a Retry-After response header indicating the retry interval.
Pick Free / Partner / Pro based on your needs. Free requires no registration; Partner is discussed with our ops team; Pro is billed through Hipiaomoney.
Free needs no key; Partner receives a slb_... key; Pro receives a gtk_... key (HMAC-signed, statelessly verifiable) after payment.
Add Authorization: Bearer <your_key> to your request header. Production endpoint: https://api.swarmlabs.tools.
Entities accumulate daily, the search index is rebuilt daily, and the OSS registry is scanned daily. Your key is isolated — data updates need no reconnection.
# 1. Full dataset, no key required (live now)
curl -s https://data.swarmlabs.tools/embodied-ai/website/api/entities.json \
| jq '.[0:3]'
# 2. Semantic search — Pro tier (requires gtk_ key)
curl -s -X POST https://api.swarmlabs.tools/v1/search/semantic \
-H "Authorization: Bearer gtk_..." \
-H "Content-Type: application/json" \
-d '{"query": "reinforcement learning safety", "sites": ["ai-safety"]}'
# 3. Partner full pull via gateway (requires slb_ key)
curl -s -H "Authorization: Bearer slb_..." \
https://api.swarmlabs.tools/v1/entities
Endpoint status: the static dataset at
data.swarmlabs.tools is live and needs no key.
The JSON API at api.swarmlabs.tools is deployed but its
custom-domain binding is pending — it will serve the endpoints above
once bound. Static access works either way.
The free tier is 60 requests per minute (not a daily or monthly quota) — normal usage rarely triggers it. If you need more headroom, upgrade to Pro for 600 requests per minute.
Entities accumulate daily (hourly batch processing). The search index is rebuilt daily and the OSS registry is scanned daily. Freshness is exposed via the lastUpdated field in /v1/domains responses.
30 science topic sites including agent-ecosystem / agritech / ai-safety / ai4science / biomed-ai / brain-science / carbon-neutral / deep-sea-tech / embodied-ai / quantum-computing / robot-parts / semiconductor, and more. Full list at /v1/domains.
arXiv / Semantic Scholar / Crossref / EuropePMC / Papers with Code / GitHub / HuggingFace. All sources are public academic or open-source data.
The ¥199 lifetime license is a one-time payment and is non-refundable. The ¥39.9 annual subscription auto-downgrades to Free when it expires without renewal. Please confirm your plan before purchasing.
The JSON API at api.swarmlabs.tools runs on Cloudflare Workers; its custom-domain binding is currently pending, so the verified-working path today is the static dataset at https://data.swarmlabs.tools/<site>/website/api/entities.json. Inside China, *.workers.dev is blocked at the network layer — use the GitHub Pages origin (data.swarmlabs.tools) or your own proxy.
Yes. The data comes from public academic and open-source sources under the MIT license. Consumers are responsible for their own downstream compliance (e.g. GDPR, CC licensing).
See the contact section below for partnership inquiries. For technical issues, use GitHub Issues at lm203688/genetech-14-sites.