NVIDIA GH200 Grace Hopper (96 GB GPU)

Hardware · NVIDIA

What AI models can the NVIDIA GH200 Grace Hopper (96 GB GPU) run?

VRAM
96.0 GB
Usable for models
96.0 GB
Memory bandwidth
4,000 GB/s
FP16 compute
133.8 TFLOPS
Released
2023
548 / 622 LLMs run Best: Qwen3.8 27B 61 / 61 Image models run Best: MiniMax H3 Image 43 / 43 Video models run Best: LTX-2.5 (22B, video + audio) 70 / 70 Decision models run Best: Qwen3 Reranker 4B

More about this hardware

MakerNVIDIA
TypeDatacenter GPU module
InterfaceNVLink

Apps that run models well on the NVIDIA GH200 Grace Hopper (96 GB GPU)

Linux

  1. vLLMThe full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.
  2. OllamaOne command with Ollama’s own build of this model.
  3. LM StudioPoint-and-click: search, download and chat in one window.
  4. SGLangThe full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.

Windows

  1. OllamaOne command with Ollama’s own build of this model.
  2. LM StudioPoint-and-click: search, download and chat in one window.
  3. llama.cppUsually the first to support new model architectures.
  4. vLLMThe full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.

For a typical model that fits in memory. Each model page picks the best app for that model on this hardware. All apps

Best LLMs for the NVIDIA GH200 Grace Hopper (96 GB GPU)

Quality Model Quant Memory Speed Runs
95Excellent quality (A+) · 95 of 100: how capable the model is at F16, whatever it runs on. Match score 97/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.8 27BAlibaba F16 61.5 GB 41.5 tok/s Runs great
94Excellent quality (A+) · 94 of 100: how capable the model is at F16, whatever it runs on. Match score 97/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 27BAlibaba F16 61.5 GB 41.5 tok/s Runs great
93Excellent quality (A+) · 93 of 100: how capable the model is at F16, whatever it runs on. Match score 97/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 35B-A3BAlibaba F16 77.5 GB 143 tok/s Runs great
93Excellent quality (A+) · 93 of 100: how capable the model is at F16, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. MiniCPM-V 4.7 35B-A3BOpenBMB F16 76.0 GB 143 tok/s Runs great
93Excellent quality (A+) · 93 of 100: how capable the model is at F16, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.5 35B-A3BAlibaba F16 77.5 GB 143 tok/s Runs great
93Excellent quality (A+) · 93 of 100: how capable the model is at F16, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.5 27BAlibaba F16 61.5 GB 41.5 tok/s Runs great
92Excellent quality (A+) · 92 of 100: how capable the model is at F16, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. Nex N2 Mini (35B-A3B)Nex-AGI F16 75.8 GB 143 tok/s Runs great
91Excellent quality (A+) · 91 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. Nemotron 3.5 Bolt 74B-A9BNVIDIA Q6_K 66.9 GB 160 tok/s Runs great
91Excellent quality (A+) · 91 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. KAT-Coder V2.5 Dev (35B-A3B)Kuaishou F16 74.9 GB 143 tok/s Runs great
91Excellent quality (A+) · 91 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Sage-A1.5 35B-A3BInternScience F16 77.5 GB 143 tok/s Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Kolibri 1 (78B-A3.5B)Aleph Alpha Q6_K 69.6 GB 216 tok/s Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Ornith 1.5 35B-A3BDeepReinforce F16 77.5 GB 143 tok/s Runs great

Best Image models for the NVIDIA GH200 Grace Hopper (96 GB GPU)

Quality Model Quant Memory Speed Runs
91Excellent quality (A+) · 91 of 100: how capable the model is at FP8, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. MiniMax H3 ImageMiniMax FP8 72.0 GB 6.3 s Runs great
86Excellent quality (A+) · 86 of 100: how capable the model is at FP16, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. Ming-Image 0.1 DesignAnt Group (inclusionAI) FP16 63.9 GB 11 s Runs great
92Excellent quality (A+) · 92 of 100: how capable the model is at FP16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. Ideogram 4Ideogram FP16 57.3 GB 17 s Runs well
89Excellent quality (A+) · 89 of 100: how capable the model is at FP16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. Qwen-Image 2.1 TurboAlibaba FP16 46.7 GB 8.1 s Runs great
92Excellent quality (A+) · 92 of 100: how capable the model is at FP16, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. Krea 2 TurboKrea FP16 37.4 GB 3.0 s Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at FP8, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. Cosmos3-Super Text2Image (64B)NVIDIA FP8 69.1 GB 25 s Runs well

Best Video models for the NVIDIA GH200 Grace Hopper (96 GB GPU)

Quality Model Quant Memory Speed Runs
87Excellent quality (A+) · 87 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. LTX-2.5 (22B, video + audio)Lightricks Q8 GGUF 49.8 GB 73 s Runs great
86Excellent quality (A+) · 86 of 100: how capable the model is at FP16, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. LTX-2.3Lightricks FP16 75.4 GB 4.6 min Runs well
82Great quality (A) · 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. LTX-2 (19B, video + audio)Lightricks FP16 75.4 GB 4.6 min Runs well
88Excellent quality (A+) · 88 of 100: how capable the model is at FP8, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. MiniMax H3 (33B, video + audio)MiniMax FP8 66.8 GB 10.9 min Runs well
86Excellent quality (A+) · 86 of 100: how capable the model is at FP16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. Wan 2.2 I2V A14BAlibaba FP16 58.0 GB 18.6 min Slow

Best Decision models for the NVIDIA GH200 Grace Hopper (96 GB GPU)

Quality Model Quant Memory Speed Runs
90Excellent quality (A+) · 90 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Qwen3 Reranker 4BAlibaba F16 13.7 GB 29 ms Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Qwen3 Reranker 8BAlibaba F16 22.1 GB 55 ms Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Skywork Reward V2 Llama 3.1 8BSkywork Q8_0 10.4 GB 91 ms Runs great
88Excellent quality (A+) · 88 of 100: how capable the model is at F16, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. zerank-2ZeroEntropy F16 13.7 GB 29 ms Runs great
86Excellent quality (A+) · 86 of 100: how capable the model is at F16, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. GLiDE 27BFastino F16 59.3 GB 91 ms Runs great
86Excellent quality (A+) · 86 of 100: how capable the model is at F16, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. Jina Reranker v3.5Jina AI F16 6.8 GB 7.7 ms Runs great
86Excellent quality (A+) · 86 of 100: how capable the model is at F16, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. Shieldstral 1.0 3BMistral AI F16 13.3 GB 28 ms Runs great
84Great quality (A) · 84 of 100: how capable the model is at F16, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. Intern-Decision 4BShanghai AI Lab F16 14.5 GB 20 ms Runs great

Best LLMs by task on this hardware Open in Model Finder

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.

Was this page helpful?