NVIDIA · Blackwell · 2025
NVIDIA B300. Rent it on demand.
From $14.30/hr per GPU on an 8-GPU machine, pay as you go with no commitment. Blackwell Ultra with 288 GB of HBM3e per GPU, 2.3 TB in an 8-GPU machine, for serving and training the largest models in FP8 and FP4.
B300 on demand
on an 8-GPU machine
- Availability
- Available now · 2 regions
- By region and size
- $14.30 to $16.77/hr
- Prepaid
- 1 mo $13.94/hr3 mo $13.59/hr6 mo $13.16/hr12 mo $12.73/hr
- Billing
- By the second on most GPUs, a few by the minute
- 8+ GPUs or a month
- Launch any size yourself, or prepay a machine for 1 to 12 months at a discount. Custom quotes on request.
The card
288 GB of HBM3e. At up to 8.0 TB/s.
- Memory
- 288 GB
- HBM3e
- Bandwidth
- up to 8.0 TB/s
- Memory to compute
- FP16 tensor
- 2250
- TFLOPS
- Per node
- 8 GPUs
- NVLink, up to 1400 W each
Workloads
What teams run on the B300. Each one opens the launch form, sized.
Launch it
Ask your agent for a B300. Or read the call it makes.
Claude Code or Codex runs it for you: it starts the instance, copies your files over and runs the job. The SDKs make the same request.
Connect your agent first
Add Runcrate's MCP server to yourself: run curl -fsSL"https://runcrate.ai/mcp.md"and follow it.
Add to Cursor
Install in VS CodeStart a B300 on Runcrate called my-gpu, copy train.py to it and run it.
create_instance { name: "my-gpu", gpu: "B300" }instance_status { id: "my-gpu" }file_upload { id: "my-gpu", content: "<train.py>", path: "/root/workspace/train.py" }ssh_execute { id: "my-gpu", command: "nohup python /root/workspace/train.py > train.log 2>&1 &" }Spec sheet
How the B300 compares. Against its closest alternatives.
Alternatives
Alternatives to the B300. Same account, different price and performance.
| GPU | Memory · FP16 tensor | Stands out | Price gap | From |
|---|---|---|---|---|
| NVIDIA B300This GPU | 288 GB HBM3e · 2250 TFLOPS | $14.30/hr | ||
| NVIDIA B200 | 180 GB HBM3e · 2250 TFLOPS | -108 GB memory | -46% | $7.67/hr |
| NVIDIA GB200 | 186 GB HBM3e · 2500 TFLOPS | -102 GB memory | Not in stock right now | |
| NVIDIA GB300 | 288 GB HBM3e · 2500 TFLOPS | Not in stock right now | ||
| AMD MI300X | 192 GB HBM3 · 1307.4 TFLOPS | -96 GB memory | Not in stock right now |
Full specs
NVIDIA B300 specifications.
Memory
- Memory
- 288 GB
- Type
- HBM3e
- Bandwidth
- up to 8.0 TB/s
Compute
- FP32
- 75 TFLOPS
- FP16 tensor
- 2250 TFLOPS
- INT8
- 187.5 TOPS
Power and form
- TDP
- up to 1400 W
- Form factor
- SXM
- Architecture
- Blackwell
- Released
- 2025
Node
- GPUs per node
- 8
- Interconnect
- NVLink
- Maker
- NVIDIA
Questions
Renting the B300. What buyers ask first.
Other GPUs
Browse every GPU by architecture.
NVIDIA Blackwell
NVIDIA Hopper
NVIDIA Ampere
NVIDIA Ada Lovelace
NVIDIA Volta