From 4c458479f9dd6951fd3c91cc52a669f8759d91fb Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Fri, 14 Aug 2026 10:00:23 -0400 Subject: [PATCH 1/9] Update choose-a-pod.mdx --- pods/choose-a-pod.mdx | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/pods/choose-a-pod.mdx b/pods/choose-a-pod.mdx index df09d5b2b..5c05cbf24 100644 --- a/pods/choose-a-pod.mdx +++ b/pods/choose-a-pod.mdx @@ -12,13 +12,16 @@ Start by identifying your primary workload type: | Workload | Recommended GPU tier | Minimum VRAM | Notes | |----------|---------------------|--------------|-------| -| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4) | 24 GB | Sufficient for most quantized models | -| **LLM inference** (30B–70B params) | High-end (A100, H100) | 48–80 GB | May require multi-GPU setup | +| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4, RTX Pro 6000 24 GB MIG*) | 24 GB | Sufficient for most quantized models | +| **LLM inference** (30B–70B params) | High-end (A100, H100, RTX Pro 6000 48 GB MIG*) | 48–80 GB | May require multi-GPU setup | | **LLM training/fine-tuning** | High-end (A100, H100) | 40–80 GB | Memory bandwidth critical | -| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4) | 16–24 GB | Benefits from fast inference | +| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4, RTX Pro 6000 24 GB MIG*) | 16–24 GB | Benefits from fast inference | | **Computer vision** | Entry to mid-range | 8–16 GB | Depends on model and batch size | | **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 | +| LLM inference (7B–13B params) | RTX Pro 6000 MIG | 24 GB | Blackwell; Secure Cloud only | +| LLM inference (30B–70B params, quantized) | RTX Pro 6000 MIG | 48 GB | Blackwell; Secure Cloud only | +| Image generation (SDXL, Flux) | RTX Pro 6000 MIG | 24 GB | Blackwell; Secure Cloud only | For a full list of available GPUs and their specifications, see [GPU types](/references/gpu-types). From 341750a1b3de866f240e9d51452ebc8c05a28a72 Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Fri, 14 Aug 2026 10:05:02 -0400 Subject: [PATCH 2/9] Update choose-a-pod.mdx --- pods/choose-a-pod.mdx | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/pods/choose-a-pod.mdx b/pods/choose-a-pod.mdx index 5c05cbf24..02448ec70 100644 --- a/pods/choose-a-pod.mdx +++ b/pods/choose-a-pod.mdx @@ -19,9 +19,9 @@ Start by identifying your primary workload type: | **Computer vision** | Entry to mid-range | 8–16 GB | Depends on model and batch size | | **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 | -| LLM inference (7B–13B params) | RTX Pro 6000 MIG | 24 GB | Blackwell; Secure Cloud only | -| LLM inference (30B–70B params, quantized) | RTX Pro 6000 MIG | 48 GB | Blackwell; Secure Cloud only | -| Image generation (SDXL, Flux) | RTX Pro 6000 MIG | 24 GB | Blackwell; Secure Cloud only | +| **LLM inference (7B–13B params)** | RTX Pro 6000 MIG | 24 GB | Blackwell; Secure Cloud only | +| **LLM inference (30B–70B params, quantized)** | RTX Pro 6000 MIG | 48 GB | Blackwell; Secure Cloud only | +| **Image generation (SDXL, Flux)** | RTX Pro 6000 MIG | 24 GB | Blackwell; Secure Cloud only | For a full list of available GPUs and their specifications, see [GPU types](/references/gpu-types). From c01ad0d75ff9574e6641e91dae3726eb265ca95e Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Fri, 14 Aug 2026 10:10:07 -0400 Subject: [PATCH 3/9] Update gpu-types.mdx --- references/gpu-types.mdx | 2 ++ 1 file changed, 2 insertions(+) diff --git a/references/gpu-types.mdx b/references/gpu-types.mdx index e250640fb..97148e7b0 100644 --- a/references/gpu-types.mdx +++ b/references/gpu-types.mdx @@ -52,6 +52,8 @@ This table lists all GPU types available on Runpod: | NVIDIA RTX A6000 | RTX A6000 | 48 | | NVIDIA RTX PRO 4500 Blackwell | RTX PRO 4500 | 32 | | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | RTX PRO 6000 MaxQ | 96 | +| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG | RTX PRO 6000 MIG | 24 | +| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG | RTX PRO 6000 MIG | 48 | | NVIDIA RTX PRO 6000 Blackwell Server Edition | RTX PRO 6000 | 96 | | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | RTX PRO 6000 WK | 96 | | Tesla V100-PCIE-16GB | Tesla V100 | 16 | From ba78393a7b58cabafbdd9b38d2f09ef8a0919d3c Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Mon, 17 Aug 2026 14:30:29 -0400 Subject: [PATCH 4/9] Update choose-a-pod.mdx --- pods/choose-a-pod.mdx | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/pods/choose-a-pod.mdx b/pods/choose-a-pod.mdx index 02448ec70..a2f3a3f11 100644 --- a/pods/choose-a-pod.mdx +++ b/pods/choose-a-pod.mdx @@ -12,19 +12,18 @@ Start by identifying your primary workload type: | Workload | Recommended GPU tier | Minimum VRAM | Notes | |----------|---------------------|--------------|-------| -| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4, RTX Pro 6000 24 GB MIG*) | 24 GB | Sufficient for most quantized models | -| **LLM inference** (30B–70B params) | High-end (A100, H100, RTX Pro 6000 48 GB MIG*) | 48–80 GB | May require multi-GPU setup | +| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4, RTX Pro 6000 24 GB MIG) | 24 GB | Sufficient for most quantized models | +| **LLM inference** (30B–70B params) | High-end (A100, H100, RTX Pro 6000 48 GB MIG) | 48–80 GB | May require multi-GPU setup | | **LLM training/fine-tuning** | High-end (A100, H100) | 40–80 GB | Memory bandwidth critical | -| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4, RTX Pro 6000 24 GB MIG*) | 16–24 GB | Benefits from fast inference | +| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4, RTX Pro 6000 24 GB MIG) | 16–24 GB | Benefits from fast inference | | **Computer vision** | Entry to mid-range | 8–16 GB | Depends on model and batch size | | **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 | -| **LLM inference (7B–13B params)** | RTX Pro 6000 MIG | 24 GB | Blackwell; Secure Cloud only | -| **LLM inference (30B–70B params, quantized)** | RTX Pro 6000 MIG | 48 GB | Blackwell; Secure Cloud only | -| **Image generation (SDXL, Flux)** | RTX Pro 6000 MIG | 24 GB | Blackwell; Secure Cloud only | For a full list of available GPUs and their specifications, see [GPU types](/references/gpu-types). +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. + ## Estimate VRAM requirements VRAM is the most common bottleneck. Use these guidelines: From 8a99319d459c26c9fe9cee55460d12c366132128 Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Tue, 18 Aug 2026 09:20:10 -0400 Subject: [PATCH 5/9] Update choose-a-pod.mdx --- pods/choose-a-pod.mdx | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/pods/choose-a-pod.mdx b/pods/choose-a-pod.mdx index a2f3a3f11..8e07fded6 100644 --- a/pods/choose-a-pod.mdx +++ b/pods/choose-a-pod.mdx @@ -12,10 +12,10 @@ Start by identifying your primary workload type: | Workload | Recommended GPU tier | Minimum VRAM | Notes | |----------|---------------------|--------------|-------| -| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4, RTX Pro 6000 24 GB MIG) | 24 GB | Sufficient for most quantized models | -| **LLM inference** (30B–70B params) | High-end (A100, H100, RTX Pro 6000 48 GB MIG) | 48–80 GB | May require multi-GPU setup | +| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4, Pro 6000 24 GB MIG) | 24 GB | Sufficient for most quantized models | +| **LLM inference** (30B–70B params) | High-end (A100, H100, Pro 6000 48 GB MIG) | 48–80 GB | May require multi-GPU setup | | **LLM training/fine-tuning** | High-end (A100, H100) | 40–80 GB | Memory bandwidth critical | -| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4, RTX Pro 6000 24 GB MIG) | 16–24 GB | Benefits from fast inference | +| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4, Pro 6000 24 GB MIG) | 16–24 GB | Benefits from fast inference | | **Computer vision** | Entry to mid-range | 8–16 GB | Depends on model and batch size | | **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 | From c78d96b306bde6e4e19b2cdb910b42b14e54371b Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Tue, 18 Aug 2026 16:09:22 -0400 Subject: [PATCH 6/9] Update gpu-types.mdx --- references/gpu-types.mdx | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/references/gpu-types.mdx b/references/gpu-types.mdx index 97148e7b0..699a6d342 100644 --- a/references/gpu-types.mdx +++ b/references/gpu-types.mdx @@ -52,8 +52,8 @@ This table lists all GPU types available on Runpod: | NVIDIA RTX A6000 | RTX A6000 | 48 | | NVIDIA RTX PRO 4500 Blackwell | RTX PRO 4500 | 32 | | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | RTX PRO 6000 MaxQ | 96 | -| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG | RTX PRO 6000 MIG | 24 | -| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG | RTX PRO 6000 MIG | 48 | +| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 1g.24gb | RTX PRO 6000 MIG | 24 | +| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 2g.48gb | RTX PRO 6000 MIG | 48 | | NVIDIA RTX PRO 6000 Blackwell Server Edition | RTX PRO 6000 | 96 | | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | RTX PRO 6000 WK | 96 | | Tesla V100-PCIE-16GB | Tesla V100 | 16 | From 21ae706ce2cf0d3e49d7a1f93848443cac750537 Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Tue, 18 Aug 2026 16:21:50 -0400 Subject: [PATCH 7/9] Update gpu-types.mdx --- references/gpu-types.mdx | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/references/gpu-types.mdx b/references/gpu-types.mdx index 699a6d342..0ee485ee6 100644 --- a/references/gpu-types.mdx +++ b/references/gpu-types.mdx @@ -52,8 +52,8 @@ This table lists all GPU types available on Runpod: | NVIDIA RTX A6000 | RTX A6000 | 48 | | NVIDIA RTX PRO 4500 Blackwell | RTX PRO 4500 | 32 | | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | RTX PRO 6000 MaxQ | 96 | -| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 1g.24gb | RTX PRO 6000 MIG | 24 | -| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 2g.48gb | RTX PRO 6000 MIG | 48 | +| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 1g.24gb | PRO 6000 MIG 24GB | 24 | +| NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 2g.48gb | PRO 6000 MIG 48GB | 48 | | NVIDIA RTX PRO 6000 Blackwell Server Edition | RTX PRO 6000 | 96 | | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | RTX PRO 6000 WK | 96 | | Tesla V100-PCIE-16GB | Tesla V100 | 16 | From 6895bd4e0fa068b2a1478347500c8297920b4da4 Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Tue, 18 Aug 2026 16:23:38 -0400 Subject: [PATCH 8/9] Update choose-a-pod.mdx --- pods/choose-a-pod.mdx | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/pods/choose-a-pod.mdx b/pods/choose-a-pod.mdx index 8e07fded6..afd956dea 100644 --- a/pods/choose-a-pod.mdx +++ b/pods/choose-a-pod.mdx @@ -12,10 +12,10 @@ Start by identifying your primary workload type: | Workload | Recommended GPU tier | Minimum VRAM | Notes | |----------|---------------------|--------------|-------| -| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4, Pro 6000 24 GB MIG) | 24 GB | Sufficient for most quantized models | -| **LLM inference** (30B–70B params) | High-end (A100, H100, Pro 6000 48 GB MIG) | 48–80 GB | May require multi-GPU setup | +| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4, Pro 6000 MIG 24 GB) | 24 GB | Sufficient for most quantized models | +| **LLM inference** (30B–70B params) | High-end (A100, H100, Pro 6000 MIG 48 GB) | 48–80 GB | May require multi-GPU setup | | **LLM training/fine-tuning** | High-end (A100, H100) | 40–80 GB | Memory bandwidth critical | -| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4, Pro 6000 24 GB MIG) | 16–24 GB | Benefits from fast inference | +| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4, Pro 6000 MIG 24 GB) | 16–24 GB | Benefits from fast inference | | **Computer vision** | Entry to mid-range | 8–16 GB | Depends on model and batch size | | **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 | From 3fe611af326b5c3ec4f59216418501601be992e3 Mon Sep 17 00:00:00 2001 From: lgunreddi Date: Tue, 18 Aug 2026 18:34:19 -0400 Subject: [PATCH 9/9] Update choose-a-pod.mdx --- pods/choose-a-pod.mdx | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/pods/choose-a-pod.mdx b/pods/choose-a-pod.mdx index afd956dea..6e3d850e8 100644 --- a/pods/choose-a-pod.mdx +++ b/pods/choose-a-pod.mdx @@ -12,17 +12,17 @@ Start by identifying your primary workload type: | Workload | Recommended GPU tier | Minimum VRAM | Notes | |----------|---------------------|--------------|-------| -| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4, Pro 6000 MIG 24 GB) | 24 GB | Sufficient for most quantized models | -| **LLM inference** (30B–70B params) | High-end (A100, H100, Pro 6000 MIG 48 GB) | 48–80 GB | May require multi-GPU setup | +| **LLM inference** (7B–13B params) | Mid-range (RTX 4090, L4, PRO 6000 MIG 24GB) | 24 GB | Sufficient for most quantized models | +| **LLM inference** (30B–70B params) | High-end (A100, H100, PRO 6000 MIG 48GB) | 48–80 GB | May require multi-GPU setup | | **LLM training/fine-tuning** | High-end (A100, H100) | 40–80 GB | Memory bandwidth critical | -| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4, Pro 6000 MIG 24 GB) | 16–24 GB | Benefits from fast inference | +| **Image generation** (SDXL, Flux) | Mid-range (RTX 4090, L4, PRO 6000 MIG 24GB) | 16–24 GB | Benefits from fast inference | | **Computer vision** | Entry to mid-range | 8–16 GB | Depends on model and batch size | | **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 | For a full list of available GPUs and their specifications, see [GPU types](/references/gpu-types). -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. +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. ## Estimate VRAM requirements