From 64e4d850a66a3be38fac7d3606468c50a9ddb0a5 Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Thu, 13 Aug 2026 16:05:52 -0400 Subject: [PATCH 1/2] Update choose-a-pod.mdx --- pods/choose-a-pod.mdx | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/pods/choose-a-pod.mdx b/pods/choose-a-pod.mdx index df09d5b2b..51b6f03bd 100644 --- a/pods/choose-a-pod.mdx +++ b/pods/choose-a-pod.mdx @@ -20,6 +20,17 @@ Start by identifying your primary workload type: | **3D rendering** | Mid-range with RT cores | 16–24 GB | RT cores accelerate ray tracing | | **Data processing** | CPU-focused or entry GPU | 8 GB+ | Prioritize CPU cores and RAM | +## Multi-Instance GPU (MIG) pods + +RTX Pro 6000 Multi-Instance GPU (MIG) slices are partitioned GPU instances with dedicated memory and compute. Available on Secure Cloud only. All MIG slices use Blackwell architecture. Verify your CUDA version and framework versions support Blackwell before deploying. + +| Workload | GPU | VRAM | Notes | +|---|---|---|---| +| LLM inference (7B–13B params) | RTX Pro 6000 MIG | 24 GB | Good alternative to RTX 4090 for inference | +| LLM inference (30B–70B params, quantized) | RTX Pro 6000 MIG | 48 GB | More VRAM than mid-range without full A100 cost | +| Image generation (SDXL, Flux) | RTX Pro 6000 MIG | 24 GB | | +| Development and prototyping | RTX Pro 6000 MIG | 24 GB | Isolated environment with dedicated resources | + For a full list of available GPUs and their specifications, see [GPU types](/references/gpu-types). ## Estimate VRAM requirements From f6bd2ee66f8c3c3a4e48ec9afcff7771f84f3c76 Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Thu, 13 Aug 2026 16:21:50 -0400 Subject: [PATCH 2/2] Update choose-a-pod.mdx --- pods/choose-a-pod.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/pods/choose-a-pod.mdx b/pods/choose-a-pod.mdx index 51b6f03bd..90bb42784 100644 --- a/pods/choose-a-pod.mdx +++ b/pods/choose-a-pod.mdx @@ -28,7 +28,7 @@ RTX Pro 6000 Multi-Instance GPU (MIG) slices are partitioned GPU instances with |---|---|---|---| | LLM inference (7B–13B params) | RTX Pro 6000 MIG | 24 GB | Good alternative to RTX 4090 for inference | | LLM inference (30B–70B params, quantized) | RTX Pro 6000 MIG | 48 GB | More VRAM than mid-range without full A100 cost | -| Image generation (SDXL, Flux) | RTX Pro 6000 MIG | 24 GB | | +| Image generation (SDXL, Flux) | RTX Pro 6000 MIG | 24 GB | Benefits from fast inference | | Development and prototyping | RTX Pro 6000 MIG | 24 GB | Isolated environment with dedicated resources | For a full list of available GPUs and their specifications, see [GPU types](/references/gpu-types).