Hyperbolic
- Categories
- Inference hosts
- Owned by
- (private)
- HQ
- US
- Website
- https://www.hyperbolic.ai
- Pricing page
- https://www.hyperbolic.ai/docs/inference/text-apis
- Fee model
- per_token+per_second_gpu
- Fees
- Serverless text models at a single $/M-token price; GPU rental (On-Demand/Reserved) is the main product, priced per GPU-hour.
- Free tier
- Basic tier $0 deposit, 60 RPM; Pro tier (>= $5 deposit) 600 RPM
- OpenAI-compatible API
- yes
- Notes
- hyperbolic.ai/pricing returns 404; marketing site now GPU-first and the docs Inference section is hidden/noindex, listing only 2 chat models. Serverless inference may be winding down.
- Sources
- https://www.hyperbolic.ai/docs/inference/text-apis https://www.hyperbolic.ai/docs/inference/overview collected 2026-09-18
LLM prices (2 models, USD per 1M tokens)
| Model | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| Llama 3.3 70B Instruct meta-llama/Llama-3.3-70B-Instruct |
$0.40 | $0.40 | — | 131k | ◐ src |
| Qwen3 Coder 480B A35B Qwen/Qwen3-Coder-480B-A35B-Instruct |
$0.40 | $0.40 | — | 262k | ◐ src |