NVIDIA B200 SXM 180GB vs NVIDIA B300 SXM 262GB

The B200 and the B300 deliver near-identical FP8 throughput (4,500 vs 4,500 TFLOPS dense). The B300 also carries 83 GB more memory (263 GB vs 180 GB).

B200: Blackwell, 2024 B300: Blackwell Ultra, 2025
Memory
180 vs 262.5 GB
83 GB more for the B300
Bandwidth
7.7 TB/s vs 8.0 TB/s
B300 moves data faster
TDP
1.0 kW vs 1.4 kW
B200 draws 400 W less
See rental pricing from $3.59/hr Download comparison sheet

Share this page

NVIDIA B200 SXM 180GB vs NVIDIA B300 SXM 262GB

A post with this link shows the card on the Image tab as its preview.

Printable version →

NVIDIA B200 SXM 180GB

Full specs →

What is the NVIDIA B200?

The B200 is a GPU from NVIDIA on the Blackwell architecture, an accelerator for AI training, inference and high-performance computing. It is an SXM module, the mezzanine format used in dense multi-GPU servers. B200 full specifications →

Can I rent the B200 in the cloud?

Yes. Cloud providers tracked by Flopper rent the B200 by the hour, and the pricing page compares their current rates and the price history side by side. B200 rental prices →

How does the B200 compare with other accelerators?

This page sets the B200 against the B300. The B200 is the older generation of the two. Both are NVIDIA parts, and both ship as an SXM module, so they fit the same class of system. Flopper also compares it head to head with the A100 SXM4 80GB and B100 SXM 192GB, across every published precision, memory and power. B200 vs A100 SXM4 80GB →

Where do the B200 figures come from?

Every B200 figure on this page traces to NVIDIA's own datasheet or specification table. How Flopper verifies specifications →

NVIDIA B300 SXM 262GB

Full specs →

What is the NVIDIA B300?

The B300 is a GPU from NVIDIA on the Blackwell Ultra architecture, an accelerator for AI training, inference and high-performance computing. It is an SXM module, the mezzanine format used in dense multi-GPU servers. B300 full specifications →

Can I rent the B300 in the cloud?

Yes. Cloud providers tracked by Flopper rent the B300 by the hour, and the pricing page compares their current rates and the price history side by side. B300 rental prices →

How does the B300 compare with other accelerators?

This page sets the B300 against the B200. The B300 is the newer generation of the two. Both are NVIDIA parts, and both ship as an SXM module, so they fit the same class of system. Flopper also compares it head to head with the H200 SXM 141GB and H100 SXM5 80GB, across every published precision, memory and power. B300 vs H200 SXM 141GB →

Where do the B300 figures come from?

Every B300 figure on this page traces to NVIDIA's own datasheet or specification table. How Flopper verifies specifications →

Specifications

B200
B300
Architecture
BlackwellBlackwell Ultra
Launch Year
20242025
Form Factor
SXMSXM
Memory
180 GB263 GB
Memory Bandwidth
7.7 TB/s8.0 TB/s
TDP
1.0 kW1.4 kW
Process Node
4nm4nm

Performance (TFLOPS)

B200
B300
37 TFLOPS 1.3 TFLOPS
75 TFLOPS 75 TFLOPS
1,125 TFLOPS
2,250 TFLOPS sparse
1,125 TFLOPS
2,250 TFLOPS sparse
2,250 TFLOPS
4,500 TFLOPS sparse
2,250 TFLOPS
4,500 TFLOPS sparse
2,250 TFLOPS
4,500 TFLOPS sparse
2,250 TFLOPS
4,500 TFLOPS sparse
4,500 TFLOPS
9,000 TFLOPS sparse
4,500 TFLOPS
9,000 TFLOPS sparse
FP6
4,500 TFLOPS
9,000 TFLOPS sparse
4,500 TFLOPS
9,000 TFLOPS sparse
9,000 TFLOPS
18,000 TFLOPS sparse
13,500 TFLOPS
18,000 TFLOPS sparse
4,500 TOPS
9,000 TOPS sparse
188 TOPS
375 TOPS sparse

Figures are dense. Where a vendor publishes a figure it sits under the dense one and never decides a winner. Hover any precision for its definition, or read the GPU glossary.

FLOPS by Precision

What actually differs

The B300 is the newer part: Blackwell Ultra (2025) against the B200's Blackwell (2024).

The B300 carries 83 GB more memory (263 GB against 180 GB), so larger models fit on one GPU before you shard. Bandwidth is 7.7 TB/s against 8.0 TB/s, which sets the ceiling on decode speed.

Power is 1.0 kW against 1.4 kW, and at FP32 the B200 does more work per watt. Weight the FP8 and FP16 rows for LLM work and the FP64 row for HPC.

New to the terms? The glossary explains FP64, FP32, TF32, BF16, FP16, FP8, FP4, INT8, Sparsity and TFLOPS. For how the generations relate, see NVIDIA architectures explained.

Frequently asked questions

What is better, the B200 or the B300?

They are close on paper: both deliver about 4,500 TFLOPS of dense FP8 throughput. Memory, bandwidth, and power are the deciding differences.

Which should you choose, the B200 or the B300?

On raw FP8 throughput the B200 and the B300 are level, so the decision rests on memory, power and price. Choose the B300 when the model has to fit on one GPU: it carries 83 GB more memory. The B200 draws 400 W less power, which matters in an air-cooled or power-capped rack.

Which has more memory, the B200 or the B300?

The B300 carries 263 GB of memory versus 180 GB for the B200. Memory bandwidth is 7.7 TB/s for the B200 and 8.0 TB/s for the B300.

How much power do the B200 and the B300 draw?

The B200 is rated at 1.0 kW TDP and the B300 at 1.4 kW. On FP32 throughput per watt, the B200 is the more efficient part.

How much do the B200 and the B300 cost to rent?

Live cloud listings tracked by Flopper price the B200 from $3.59 per GPU-hour across 25 providers, and the B300 from $4.08 per GPU-hour across 19 providers. The rental section on this page links to every provider, rate and the price history for each GPU.

Where to Rent

NVIDIA B200 SXM 180GB

Rent from

$3.59 / GPU-hr

Lowest of 35 live listings across 25 providers · median $6.92

See all B200 rental prices →

Price history and every provider on the pricing page. Some links are affiliate links.

NVIDIA B300 SXM 262GB

Rent from

$4.08 / GPU-hr

Lowest of 24 live listings across 19 providers · median $7.92

See all B300 rental prices →

Price history and every provider on the pricing page. Some links are affiliate links.

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.

© 2026 Flopper.io - Compare the GPUs Powering AI