GPU cloud
Modal
/modal · New York, USA · est. 2021
Data: mediumVisit provider ↗
Key facts
- Modal is a gpu cloud headquartered in New York, USA, founded 2021.
- Uses its own native API dialect rather than an OpenAI-compatible endpoint.
- Billed by GPU-time rather than per token, so no per-token blended price is listed.
- Reference latency 700 ms TTFT at 100 tok/s throughput, with 99.7% reference uptime.
- Editorial reputation index 85/100 from 300 weighted signals; 90-day trend improving.
Specification
- Category
- GPU cloud
- Headquarters
- New York, USA
- Founded
- 2021
- Model access
- Open weights
- Flagship / served
- Serverless GPU runtime
- Reference open model
- Bring-your-own; billed by GPU-second
- OpenAI-compatible API
- No (native API)
- Blended $/1M
- — seed
Strengths & trade-offs
Strength
Code-first serverless GPUs; full control of the serving stack.
Trade-off
You build the inference layer; not a turnkey token API.
Profile reviewed 2026-07-10. Confidence for this listing:medium — Figure based on public sources but not re-confirmed on the exact review date; may drift.Not affiliated with Modal. ← Back to directory