Hyperbolic
Hyperbolic Labs · hyperbolic.ai · Last checked:
- Serverless per-token API
- GPU rental (on-demand and reserved)
Company status
Active, privately held.
Sources 1, 2, 3 · checked 2026-11-08
Capabilities
- Serverless inference for open-weight text and vision-language models through an OpenAI-compatible chat completions API
- Billing per 1M tokens from a prepaid credit balance
- H100, H200, and B200 GPU instances on demand or reserved, plus managed Kubernetes, for self-hosting models
Sources 1, 2, 3 · checked 2026-11-08
Who it tends to fit
Pricing
Published per-token pricing on https://www.hyperbolic.ai/docs/inference/overview (USD per 1M tokens, dated October 21, 2026; for example glm-5.3-flash listed at $0.15 input and $0.50 output).
Sources 1, 2, 3 · checked 2026-11-08
Published certifications
Limitations to check
- SOC 2 Type II not yet complete per the vendor's security page
- Vendor states it is not HIPAA-certified; do not send PHI without consulting them
- Short serverless model list; image, TTS, and base completion models from the previous API are not available
- Free-tier accounts need phone verification or a Pro deposit to use inference
Sources 1, 2, 3 · checked 2026-11-08
Sources
- Inference overview: OpenAI-compatible chat completions; billed per million tokens from credit balance; model price table as of October 21, 2026www.hyperbolic.ai/docs/inference/overviewChecked 2026-11-08
- Security and compliance: SOC 2 Type I completed, Type II underway; not currently HIPAA-certifiedhyperbolic.ai/docs/general/security-complianceChecked 2026-11-08
- Homepage: H100, H200, B200 on demand or reserved; managed Kubernetes; inferencewww.hyperbolic.ai/Checked 2026-11-08
Report a correction: [email protected]. Facts are re-checked at least every 90 days.