Owned and curated capacity
Managed Fleet capacity for steady production
TensorNova Fleet is the public owned-capacity rail for repeatable workloads. Enterprise sits above it with larger allocations, priority Fleet access, and direct machine or cluster admin access.
TensorNova performance
Steady public capacity for repeatable production jobs
Run production workloads on TensorNova-owned inventory with clearer capacity signals, InfiniBand-backed networking, and TensorNova-managed deployment. Enterprise sits above this rail when the job needs dedicated infrastructure access.
See how Enterprise changes access →Delivery model
Managed runtime
Best fit
Steady production
Regions
3 active
TN-H100-GEN
NVIDIA H100
US East
VRAM
80 GB
Nodes
64
Network
800 Gbps
SLA tier
Business
- • Managed runtime
- • Higher baseline capacity
- • Instant clusters
Fleet capacity
$3.39/hr from
Request FleetTN-H200-SCL
NVIDIA H200
APAC Singapore
VRAM
141 GB
Nodes
32
Network
800 Gbps
SLA tier
Business
- • High-memory fleet
- • Managed runtime
- • Reservation path
Fleet capacity
$4.02/hr from
Request FleetTN-B200-PRI
NVIDIA B200
APAC Singapore
VRAM
192 GB
Nodes
24
Network
800 Gbps
SLA tier
Business
- • Blackwell generation
- • Managed deployment
- • Enterprise priority available
Fleet capacity
$6.32/hr from
Request FleetVerified public marketplace
Community hosts are benchmarked, identity-verified, and continuously scored before TensorNova allows them behind the public control plane.
Fleet as steady capacity
TensorNova Fleet is public managed capacity for repeatable workloads. It sits above limited community supply and gives teams a steadier operating rail.
Enterprise gets the direct path
Enterprise contracts add larger allocations, prioritized Fleet access, and direct machine or cluster admin access when managed runtime is no longer enough.
Renter request
Request managed Fleet capacity
Share region, GPU target, and expected daily load. TensorNova will confirm whether public Fleet fits before moving you into premium enterprise allocation.
Enterprise path
Need larger blocks or dedicated admin access?
Public Fleet stays inside TensorNova-managed runtime. Enterprise is where teams buy larger capacity blocks, prioritize Fleet allocation, and move into direct machine or cluster admin access.