Skip to content

NIM GPU bundle validation

This page summarizes which NIM models are regression-tested or estimated on each standard GPU bundle across major clouds and air-gap EKS, where letters in the EKS, AKS, and OCP columns indicate the minimum bundle size defined in NIM GPU resource bundles.

Legend: - ✓ Validated – regression-tested on the cloud with the listed bundle. - □ Estimated – expected to work but not yet regression-tested.

モデル バージョン EKS (AWS) AKS (Azure) OCP (Oracle) EKS AirGap OKE H100
genmol 1.0.0 ✓ M ✓ M ✓ XL - ✓ XL
molmim 1.0.0 ✓ M ✓ M ✓ XL - ✓ XL
proteinmpnn 1.0.2 ✓ M ✓ M ✓ XL ✓ M ✓ XL
rfdiffusion 2.2.0 ✓ L ✓ L ✓ XL ✓ L ✓ XL
arctic-embed-l 1.0.1 ✓ M ✓ M ✓ XL - ✓ XL
nv-embedqa-e5-v5 1.6.0 ✓ M ✓ M ✓ XL ✓ M □ XL
nv-embedqa-e5-v5-pb24h2 1.2.3 ✓ M ✓ M ✓ XL ✓ M ✓ XL
nv-embedqa-mistral-7b-v2 1.0.1 ✓ L ✓ L ✓ XL - ✓ XL
nvclip 2.0.0 ✓ M ✓ M ✓ XL - □ XL
llama-3.1-nemoguard-8b-content-safety 1.0.0 □ L ✓ XL ✓ L ✓ XL - □ 2XL
llama-3.1-nemoguard-8b-topic-control 1.0.0 □ L ✓ XL ✓ L ✓ XL - □ 2XL
nemoguard-jailbreak-detect 1.0.0 ✓ M ✓ M ✓ XL - ✓ XL
codellama-13b-instruct 1.2.2 ✓ XL □ 2XL ✓ 2XL - ✓ 2XL
codellama-34b-instruct 1.2.2 ✓ XL □ 2XL ✓ 2XL - ✓ 2XL
codellama-70b-instruct 1.2.2 ✓ 2XL □ 3XL ✓ 3XL - ✓ 3XL
deepseek-r1-distill-llama-8b 1.5.2 ✓ L ✓ L ✓ XL ✓ L □ XL
deepseek-r1-distill-qwen-7b 1.1.0 ✓ M ✓ M ✓ XL ✓ M □ XL
gemma-2-2b-instruct 1.4.0 ✓ M ✓ M ✓ XL - □ XL
gemma-2-9b-it 1.4.0 ✓ XL □ XL ✓ XL - □ XL
llama-2-13b-chat 1.0.3 ✓ L □ L ✓ XL - ✓ XL
llama-2-70b-chat 1.0.3 □ 3XL □ 3XL ✓ 3XL - ✓ 3XL
llama-2-7b-chat 1.0.3 ✓ L □ L ✓ XL - ✓ XL
llama-3-sqlcoder-8b 1.2.3 ✓ M □ M ✓ XL - ✓ XL
llama-3-swallow-70b-instruct-v0.1 1.1.2 □ 2XL □ 3XL ✓ 3XL - ✓ 3XL
llama-3-taiwan-70b-instruct 1.1.2 □ 2XL □ 3XL ✓ 3XL - ✓ 3XL
llama-3.1-70b-instruct 1.8.4 ✓ 2XL □ 2XL □ 2XL ✓ 3XL ✓ 2XL ✓ 2XL
llama-3.1-8b-instruct 1.8.4 ✓ L ✓ L ✓ XL ✓ L ✓ XL
llama-3.1-nemotron-70b-instruct 1.2.3 ✓ 3XL □ 3XL ✓ 3XL - ✓ 3XL
llama-3.1-swallow-70b-instruct-v0.1 1.3.2 ✓ 3XL □ 3XL □ 3XL ✓ 4XL - □ 3XL
llama-3.2-11b-vision-instruct 1.1.1 ✓ XL □ XL ✓ XL - ✓ XL
llama-3.2-nv-embedqa-1b-v2 1.6.0 ✓ M ✓ M ✓ XL ✓ M □ XL
llama-3.3-70b-instruct 1.8.5 ✓ 3XL □ 4XL ✓ 4XL ✓ 3XL ✓ 4XL
llama3-70b-instruct 1.0.3 □ 3XL □ 3XL ✓ 3XL - ✓ 3XL
llama3-8b-instruct 1.0.3 ✓ M □ M ✓ XL - ✓ XL
mistral-7b-instruct-v0.3 1.3.0 ✓ L ✓ L ✓ XL ✓ L □ XL
mistral-nemo-12b-instruct 1.2.2 ✓ XL □ XL ✓ XL - ✓ XL
mistral-nemo-minitron-8b-8k-instruct 1.2.3 ✓ L □ L ✓ XL - ✓ XL
mixtral-8x7b-instruct-v01 1.3.0 ✓ 2XL □ 2XL ✓ 2XL ✓ 2XL □ 2XL
phi-3-mini-4k-instruct 1.2.3 ✓ M □ M ✓ XL - ✓ XL
qwen-2.5-7b-instruct 1.0.0 ✓ M ✓ M ✓ XL - □ XL
llama-3.2-nv-rerankqa-1b-v2 1.5.0 ✓ M ✓ M ✓ XL ✓ M □ XL
nv-rerankqa-mistral-4b-v3 1.0.2 ✓ M ✓ M ✓ XL - ✓ XL
paddleocr 1.3.0 ✓ S ✓ S ✓ XL - □ XL
llama-3.2-1b-instruct 1.8.5 ✓ M ✓ M ✓ XL ✓ M ✓ XL
nemoretriever-graphic-elements-v1 1.3.0 ✓ M ✓ M ✓ XL ✓ M □ XL
nemoretriever-page-elements-v2 1.3.0 ✓ M ✓ M ✓ XL ✓ M □ XL
nemoretriever-parse 1.2.0 ✓ M □ M ✓ XL ✓ M ✓ XL
nemoretriever-table-structure-v1 1.3.0 ✓ M ✓ M ✓ XL ✓ M □ XL
deepseek-r1-distill-qwen-14b 1.1.0 ✓ L □ L ✓ XL ✓ L □ XL
llama-3.1-8b-instruct-pb24h2 1.3.6 ✓ L ✓ L ✓ XL ✓ L □ XL
llama-3.1-nemotron-nano-8b-v1 1.8.3 ✓ L ✓ L ✓ XL ✓ L ✓ XL
llama-3.2-3b-instruct 1.8.4 ✓ L ✓ L ✓ XL ✓ L ✓ XL
openfold2 1.0.0 □ XL □ XL ✓ XL - ✓ XL
cosmos-predict1-7b-text2world 1.0.0 - - □ 2XL - ✓ 2XL
cosmos-predict1-7b-video2world 1.0.0 - - □ 2XL - ✓ 2XL
deepseek-r1-distill-qwen-32b 1.1.0 ✓ XL □ 2XL ✓ 2XL ✓ XL ✓ 2XL
llama-3.1-70b-instruct-pb24h2 1.3.6 ✓ 2XL □ 2XL □ 2XL ✓ 3XL ✓ 2XL □ 2XL
llama-3.2-90b-vision-instruct 1.1.1 ✓ 3XL □ 3XL ✓ 3XL - ✓ 3XL
llama-3.3-nemotron-super-49b-v1 1.8.6 ✓ 2XL □ 3XL ✓ 3XL ✓ 2XL ✓ 3XL
starcoder2-7b 1.8.1 ✓ XL □ 2XL ✓ 2XL ✓ XL ✓ 2XL
llama-3.1-nemotron-ultra-253b-v1 1.8.4 □ 3XL □ 4XL □ 4XL - ✓ 4XL
mixtral-8x22b-instruct-v01 1.2.2 □ 4XL □ 4XL ✓ 4XL - ✓ 4XL
cuopt 25.02.01 ✓ M ✓ M ✓ XL ✓ M □ 2XL
llama-3.3-nemotron-super-49b-v1.5 1.12.0 ✓ 2XL - ✓ 2XL - ✓ 2XL
llama-4-scout-17b-16e-instruct 1.3.2 □ 3XL - ✓ 3XL - ✓ 3XL
gpt-oss-20b 1.12.3 ✓ M ✓ L ✓ XL - ✓ XL - ✓ XL
gpt-oss-120b 1.12.3 ✓ XL ✓ 2XL - ✓ XL - □ XL ✓ 2XL
qwen3-next-80b-a3b-thinking 1.0.0 - - □ 3XL - ✓ 3XL
nvidia-nemotron-nano-9b-v2 1.12.2 □ L □ L ✓ XL - ✓ XL
qwen3-32b 1.0.0 - - □ 2XL - ✓ XL