Skip to content

Commit 3bac255

Browse files
bene2k1jcirinosclwy
bene2k1
and
jcirinosclwy
authored
fix(gpu): fix table (#5928)
* fix(gpu): fix table * Update pages/gpu/reference-content/choosing-gpu-instance-type.mdx Co-authored-by: Jessica <113192637+jcirinosclwy@users.noreply.github.com> --------- Co-authored-by: Jessica <113192637+jcirinosclwy@users.noreply.github.com>
1 parent 10ebffc commit 3bac255

File tree

1 file changed

+18
-19
lines changed

1 file changed

+18
-19
lines changed

‎pages/gpu/reference-content/choosing-gpu-instance-type.mdx‎

Lines changed: 18 additions & 19 deletions
Original file line numberDiff line numberDiff line change
@@ -40,23 +40,6 @@ Remember that there is no one-size-fits-all answer, and the right GPU Instance t
4040

4141
### Scaleway GPU Instances types overview
4242

43-
| | **[RENDER-S](https://www.scaleway.com/en/gpu-render-instances/)** | **[H100-1-80G](https://www.scaleway.com/en/h100-pcie-try-it-now/)** | **[H100-2-80G](https://www.scaleway.com/en/h100-pcie-try-it-now/)** |
44-
|---------------------------------------------------------------------|-------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------|
45-
| GPU Type | 1x [P100](https://www.nvidia.com/en-us/data-center/tesla-p100/) PCIe3 | 1x [H100](https://resources.nvidia.com/en-us-tensor-core/nvidia-tensor-core-gpu-datasheet) PCIe5 | 2x [H100](https://resources.nvidia.com/en-us-tensor-core/nvidia-tensor-core-gpu-datasheet) PCIe5 |
46-
| NVIDIA architecture | Pascal 2016 | Hopper 2022 | Hopper 2022 |
47-
| Tensor Cores | N/A | Yes | Yes |
48-
| Performance (training in FP16 Tensor Cores) | (No Tensor Cores : 9,3 TFLOPS FP32) | 1513 TFLOPS | 2x 1513 TFLOPS |
49-
| VRAM | 16 GB CoWoS HBM2 (Memory bandwidth: 732 GB/s) | 80 GB HBM2E (Memory bandwidth: 2TB/s) | 2x80 GB HBM2E (Memory bandwidth: 2TB/s) |
50-
| CPU Type | Intel Xeon Gold 6148 (2.4 GHz) | AMD EPYC™ 9334 (2.7GHz) | AMD EPYC™ 9334 (2.7GHz) |
51-
| vCPUs | 10 | 24 | 48 |
52-
| RAM | 42 GB DDR3 | 240 GB DDR5 | 480 GB DDR5 |
53-
| Storage | Block/Local | Block | Block |
54-
| [Scratch Storage](/gpu/how-to/use-scratch-storage-h100-instances/) | No | Yes (3 TB NVMe) | Yes (6 TB NVMe) |
55-
| [MIG compatibility](/gpu/how-to/use-nvidia-mig-technology/) | No | Yes | Yes |
56-
| Bandwidth | 1 Gbps | 10 Gbps | 20 Gbps |
57-
| Better used for | Image / Video encoding (4K) | 7B LLM Fine-Tuning / Inference | 70B LLM Fine-Tuning / Inference |
58-
| What they are not made for | Large models (especially LLM) | Graphic or video encoding use cases | Graphic or video encoding use cases |
59-
6043
| | **[B300-SXM-8-288G](https://www.scaleway.com/en/b300-sxm/)** |
6144
|--------------------------------------------------------------------|----------------------------------------------------------------------------|
6245
| GPU type | 8x [B300-SXM](https://www.nvidia.com/en-us/data-center/dgx-b300/) |
@@ -73,8 +56,25 @@ Remember that there is no one-size-fits-all answer, and the right GPU Instance t
7356
| Bandwidth | 20 Gbps |
7457
| Network technology | [NVLink](/gpu/reference-content/understanding-nvidia-nvlink/) |
7558
| Better used for | Deploying large-scale AI model training and inference workloads — especially large LLMs, multimodal AI, or heavy HPC tasks |
76-
7759
| What they are not made for | Real-time graphics, video editing or game-graphics workloads |
60+
61+
| | **[RENDER-S](https://www.scaleway.com/en/gpu-render-instances/)** | **[H100-1-80G](https://www.scaleway.com/en/h100-pcie-try-it-now/)** | **[H100-2-80G](https://www.scaleway.com/en/h100-pcie-try-it-now/)** |
62+
|---------------------------------------------------------------------|-------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------|
63+
| GPU Type | 1x [P100](https://www.nvidia.com/en-us/data-center/tesla-p100/) PCIe3 | 1x [H100](https://resources.nvidia.com/en-us-tensor-core/nvidia-tensor-core-gpu-datasheet) PCIe5 | 2x [H100](https://resources.nvidia.com/en-us-tensor-core/nvidia-tensor-core-gpu-datasheet) PCIe5 |
64+
| NVIDIA architecture | Pascal 2016 | Hopper 2022 | Hopper 2022 |
65+
| Tensor Cores | N/A | Yes | Yes |
66+
| Performance (training in FP16 Tensor Cores) | (No Tensor Cores : 9,3 TFLOPS FP32) | 1513 TFLOPS | 2x 1513 TFLOPS |
67+
| VRAM | 16 GB CoWoS HBM2 (Memory bandwidth: 732 GB/s) | 80 GB HBM2E (Memory bandwidth: 2 TB/s) | 2x80 GB HBM2E (Memory bandwidth: 2 TB/s) |
68+
| CPU Type | Intel Xeon Gold 6148 (2.4 GHz) | AMD EPYC™ 9334 (2.7GHz) | AMD EPYC™ 9334 (2.7GHz) |
69+
| vCPUs | 10 | 24 | 48 |
70+
| RAM | 42 GB DDR3 | 240 GB DDR5 | 480 GB DDR5 |
71+
| Storage | Block/Local | Block | Block |
72+
| [Scratch Storage](/gpu/how-to/use-scratch-storage-h100-instances/) | No | Yes (3 TB NVMe) | Yes (6 TB NVMe) |
73+
| [MIG compatibility](/gpu/how-to/use-nvidia-mig-technology/) | No | Yes | Yes |
74+
| Bandwidth | 1 Gbps | 10 Gbps | 20 Gbps |
75+
| Better used for | Image / Video encoding (4K) | 7B LLM Fine-Tuning / Inference | 70B LLM Fine-Tuning / Inference |
76+
| What they are not made for | Large models (especially LLM) | Graphic or video encoding use cases | Graphic or video encoding use cases |
77+
7878
| | **[H100-SXM-2-80G](https://www.scaleway.com/en/h100-pcie-try-it-now/)** | **[H100-SXM-4-80G](https://www.scaleway.com/en/h100-pcie-try-it-now/)** | **[H100-SXM-8-80G](https://www.scaleway.com/en/h100-pcie-try-it-now/)** |
7979
|--------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|-------------------------------------------------------------------|
8080
| GPU Type | 2x [H100-SXM](https://www.nvidia.com/en-us/data-center/h100/) SXM | 4x [H100-SXM](https://www.nvidia.com/en-us/data-center/h100/) SXM | 8x [H100-SXM](https://www.nvidia.com/en-us/data-center/h100/) SXM |
@@ -125,7 +125,6 @@ Remember that there is no one-size-fits-all answer, and the right GPU Instance t
125125
| [MIG compatibility](/gpu/how-to/use-nvidia-mig-technology/) | No | No | No | No |
126126
| Bandwidth | 2,5 Gbps | 5 Gbps | 10 Gbps | 20 Gbps |
127127
| Use cases | GenAI (Image/Video) | GenAI (Image/Video) | 7B Text-to-image model fine-tuning / Inference | 70B text-to-image model fine-tuning / Inference |
128-
| What they are not made for | | | | |
129128

130129
<Message type="note">
131130
The service level objective (SLO) for all GPU Instance types is 99.5% availability. [Read the SLA](https://www.scaleway.com/en/virtual-instances/sla/).

0 commit comments

Comments
 (0)