What's the price difference between NVIDIA B300 and NVIDIA A100 40GB?

On Runcrate today, NVIDIA B300 starts at $10.40/hr and NVIDIA A100 40GB starts at $1.99/hr; NVIDIA A100 40GB is cheaper by 81%.

Can I deploy NVIDIA B300 or NVIDIA A100 40GB on Runcrate today?

NVIDIA B300 is on demand in 1 region from $10.40/hr, billed by the second on most GPUs. NVIDIA A100 40GB is on demand in 6 regions from $1.99/hr, billed by the second on most GPUs.

How much VRAM does each GPU have?

NVIDIA B300 has 288GB HBM3e. NVIDIA A100 40GB has 40GB HBM2.

B300 or A100 40GB: which holds a 70B model in 8-bit weights?

It needs 70 GB of 8-bit weights. On the B300: 1 GPU, 218 GB spare. On the A100 40GB: 2 GPUs (80 GB). Pick the B300: fits on one GPU.

GPU comparison

B300 vs A100 40GB.

The B300 is ahead on memory (288 GB vs 40 GB) and speed (7.2x the FP16 throughput and up to 5.1x the memory bandwidth). The A100 40GB makes sense when your model fits in 40 GB and its hourly price is lower.

Launch A100 40GB Reserve a cluster

Rent either on Runcrate

B300from $10.40/hr
$0.036 per GB/hr$0.46 per 100 TFLOPS/hr
Launch B300
A100 40GBCheaperfrom $1.99/hr
$0.050 per GB/hr$0.64 per 100 TFLOPS/hr
Launch A100 40GB

Live hourly price, then per GB of memory and per 100 FP16 TFLOPS.

Which is better for

Which to pick for each job. Memory fit and throughput, from the datasheets.

JobB300A100 40GBPick

Hold a 70B model in 8-bit weights
70 GB of 8-bit weights
B300: 1 GPU, 218 GB spare
A100 40GB: 2 GPUs (80 GB)
B300. Fits on one GPU.
Serve an 8B model in 16-bit weights
16 GB of 16-bit weights
B300: 1 GPU, 272 GB spare
A100 40GB: 1 GPU, 24 GB spare
B300. Same GPU count, up to 5.1x the memory bandwidth.
Fine-tune a 70B model with QLoRA
About 46 GB with 4-bit weights
B300: 1 GPU, 242 GB spare
A100 40GB: 2 GPUs (80 GB)
B300. Fits on one GPU.
Train on a full node
FP16 tensor throughput per node
B300: 8x, 18,000 TFLOPS, NVLink
A100 40GB: 8x, 2,496 TFLOPS
B300. 7.2x the FP16 throughput per GPU.
Throughput per watt
FP16 TFLOPS per 100 W of TDP
B300: Varies by card
A100 40GB: Varies by card
Either. Varies by card.

Compare another pair

Blackwell

NVIDIA B300

from$10.40/hr

Available now · 1 region

Memory: 288 GB HBM3e
FP16 tensor: 2250 TFLOPS
Bandwidth: up to 8.0 TB/s

Launch B300 Reserve a cluster

Full pricing details

Ampere

NVIDIA A100 40GB

from$1.99/hrCheaper

Available now · 6 regions

Memory: 40 GB HBM2
FP16 tensor: 312 TFLOPS
Bandwidth: 1.6 TB/s

Launch A100 40GB Reserve a cluster

Full pricing details

At a glance

Spec by spec. The longer bar is the better number.

Memory

NVIDIA B300288 GB

NVIDIA A100 40GB40 GB

Memory bandwidth

NVIDIA B300up to 8.0 TB/s

NVIDIA A100 40GB1.6 TB/s

FP16 tensor

NVIDIA B3002250 TFLOPS

NVIDIA A100 40GB312 TFLOPS

Hourly price, from (live)

NVIDIA B300$10.40/hr

NVIDIA A100 40GB$1.99/hr

Full specs

The complete spec sheet.

Spec	NVIDIA B300	NVIDIA A100 40GB
Architecture	Blackwell	Ampere
Memory	288 GB	40 GB
Memory type	HBM3e	HBM2
Memory bandwidth	up to 8.0 TB/s	1.6 TB/s
FP32	75 TFLOPS	19.5 TFLOPS
FP16 tensor	2250 TFLOPS	312 TFLOPS
INT8	187.5 TOPS	624 TOPS
TDP	up to 1400 W	up to 400 W
GPUs per node	8	8
Hourly price (live)	from $10.40/hr	from $1.99/hr

Questions

B300 or A100 40GB. What buyers ask first.

Other comparisons