Lyceum
vs
Packet·ai
What's good about each
- Data center cards from L40S to Blackwell B300, 1 to 8 per VM
- European datacenters, and prompts and outputs are not stored
- Per-second billing on compute, per-token on inference
- OpenAI-compatible API that plugs into coding agents
- Dynamic tier: shared GPUs at about half the dedicated rate
- Free, unlimited data egress
- Single GPUs from RTX 4090 to B200
- S3-compatible object storage and OpenAI-compatible inference API
Price comparison
Closest matching configuration on each side, monthly list price in USD.
| Service | Lyceum | Packet·ai | Difference |
|---|---|---|---|
| Block storage | -- Not listed | $7.20 /mo 100 GB | |
| Object storage | -- Not listed | $18.00 /mo 1 TB | |
| Egress | Free and unlimited | Free and unlimited |
Heads up: Figures take each provider's closest plan to a common setup, and vary by region and usage. Hover a service for the reference configuration. Verify before provisioning. More on how we price.
GPUs
Lyceum and Packet·ai each offer 7 GPU models. Cheapest listed price per GPU per hour; GPUs both providers sell come first.
| GPU | Lyceum | Packet·ai | Difference |
|---|---|---|---|
B300
9 configs · 1 config
|
$6.39 /GPU/hr Spot | -- On request | |
|
H100
9 configs · 2 configs
|
$2.69 /GPU/hr Spot | -- On request | |
|
B200
9 configs · 6 configs
|
$9.99 /GPU/hr On-demand · sold out | $4.25 /GPU/hr On-demand | 57% cheaper on Packet·ai |
|
RTX PRO 6000
8 configs · 3 configs
|
$2.99 /GPU/hr On-demand | $0.76 /GPU/hr On-demand | 75% cheaper on Packet·ai |
|
H200
9 configs · 1 config
|
$3.59 /GPU/hr Spot · sold out | -- On request | |
|
A100
9 configs · 1 config
|
$2.59 /GPU/hr On-demand · sold out | $0.89 /GPU/hr On-demand | 66% cheaper on Packet·ai |
|
L40S
9 configs
|
$2.19 /GPU/hr On-demand | -- | Lyceum only |
|
RTX 4090
1 config
|
-- | $0.39 /GPU/hr On-demand | Packet·ai only |
Lyceum bills on-demand and spot; Packet·ai bills on-demand and reserved (1 months). The line under each price names the billing type shown.
Services
Managed services each provider offers.
| Service | Lyceum | Packet·ai |
|---|---|---|
| Block Storage | ||
| GPU-powered Servers | ||
| Managed Containers | ||
| Object Storage | ||
| Python Notebooks |
Compliance
Frameworks each provider states on its own site, each linked to its source.
Heads up: Each entry links to the provider's own statement. An attestation can cover one tier, one region or the platform alone rather than every service. Verify the scope before you rely on it. A dash means we found nothing to link to, not that the provider lacks it.
Company details
| Fact | Lyceum | Packet·ai |
|---|---|---|
| Founded | 2025 | 2025 |
| Headquarters | Germany | USA |
| Locations | 2 | 12 |
| Example customers | Enpal, Apheris, Edgeless Systems, Macrocosmos, SONA | -- |
| Website | lyceum.technology | packet.ai |
Alternatives to consider
Compare Lyceum and Packet·ai against other cloud providers.
Our data for Lyceum was last updated on Nov. 11, 2026, and for Packet·ai on Nov. 11, 2026.