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

Specifications

Maker	NVIDIA
Type	Datacenter GPU module
Video memory (VRAM)	96.0 GB
Usable for AI models	96.0 GB
Memory bandwidth	4,000 GB/s
FP16 compute	133.8 TFLOPS
Interface	NVLink
Released	2023

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

Grade · points	Model	Quant	Memory	Speed	Fit
97 ptsExcellent (S) · 97 of 100 points: Model quality 43.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.8/15	Qwen3.8 27BAlibaba	`F16`	61.5 GB	41.5 tok/s	Runs well
97 ptsExcellent (S) · 97 of 100 points: Model quality 42.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.8/15	Qwen3.6 27BAlibaba	`F16`	61.5 GB	41.5 tok/s	Runs well
97 ptsExcellent (S) · 97 of 100 points: Model quality 42.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15	Qwen3.6 35B-A3BAlibaba	`F16`	77.5 GB	192 tok/s	Runs well
96 ptsExcellent (S) · 96 of 100 points: Model quality 41.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15	Qwen3.5 35B-A3BAlibaba	`F16`	77.5 GB	192 tok/s	Runs well
96 ptsExcellent (S) · 96 of 100 points: Model quality 41.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.8/15	Qwen3.5 27BAlibaba	`F16`	61.5 GB	41.5 tok/s	Runs well
95 ptsExcellent (S) · 95 of 100 points: Model quality 40.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15	Ornith 1.5 35B-A3BDeepReinforce	`F16`	77.5 GB	192 tok/s	Runs well
95 ptsExcellent (S) · 95 of 100 points: Model quality 41.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 13.8/15	Gemma 4 26B-A4BGoogle	`F16`	57.4 GB	152 tok/s	Runs well
95 ptsExcellent (S) · 95 of 100 points: Model quality 40.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15	TMax 27BAi2	`F16`	60.7 GB	42.1 tok/s	Runs well
95 ptsExcellent (S) · 95 of 100 points: Model quality 40.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15	Apodex 1.1 Mini (35B-A3B)Apodex	`F16`	77.5 GB	192 tok/s	Runs well
95 ptsExcellent (S) · 95 of 100 points: Model quality 40.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15	Ornith 1.0 35B-A3BDeepReinforce	`F16`	77.5 GB	192 tok/s	Runs well
94 ptsExcellent (S) · 94 of 100 points: Model quality 39.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15	GLM-4.7 Flash (30B-A3B)Zhipu AI	`F16`	68.3 GB	192 tok/s	Runs well
94 ptsExcellent (S) · 94 of 100 points: Model quality 39.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15	Fara 1.5 27BMicrosoft	`F16`	60.7 GB	42.1 tok/s	Runs well

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

Grade · points	Model	Precision	Memory	Time per output	Fit
89 ptsExcellent (S) · 89 of 100 points: Model quality 40.6/45, Memory fit 25.0/25, Speed 10.0/15, Memory use 14.3/15	FLUX.2 [dev]Black Forest Labs	`FP8`	59.7 GB	23 s	Runs well
89 ptsExcellent (S) · 89 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15	HiDream-I1 FastHiDream.ai	`FP16`	64.4 GB	7.0 s	Runs well
86 ptsExcellent (S) · 86 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.9/15	FLUX.2 [klein] 9BBlack Forest Labs	`FP16`	37.1 GB	1.3 s	Runs well
86 ptsExcellent (S) · 86 of 100 points: Model quality 40.5/45, Memory fit 25.0/25, Speed 6.0/15, Memory use 15.0/15	Qwen-Image 2512Alibaba	`FP16`	63.8 GB	63 s	Runs well
86 ptsExcellent (S) · 86 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 8.5/15, Memory use 15.0/15	HiDream-I1 FullHiDream.ai	`FP16`	64.4 GB	31 s	Runs well
85 ptsExcellent (S) · 85 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.7/15	FLUX.1 Krea [dev]Black Forest Labs	`FP16`	36.4 GB	8.4 s	Runs well

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

Grade · points	Model	Precision	Memory	Time per output	Fit
90 ptsExcellent (S) · 90 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 12.1/15, Memory use 15.0/15	LTX-2.3Lightricks	`FP16`	75.4 GB	4.6 min	Runs well
88 ptsExcellent (S) · 88 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 12.1/15, Memory use 15.0/15	LTX-2 (19B, video + audio)Lightricks	`FP16`	75.4 GB	4.6 min	Runs well
83 ptsGreat (A) · 83 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 6.0/15, Memory use 13.9/15	Wan 2.2 I2V A14BAlibaba	`FP16`	58.0 GB	18.6 min	Runs well
83 ptsGreat (A) · 83 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 6.0/15, Memory use 13.9/15	Wan 2.2 T2V A14BAlibaba	`FP16`	58.0 GB	18.6 min	Runs well
82 ptsGreat (A) · 82 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 12.1/15, Memory use 11.9/15	FramePack I2V (HunyuanVideo 13B)lllyasviel	`FP16`	49.6 GB	4.6 min	Runs well

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.