Can Llama 4 Maverick 17B 128E run on NVIDIA H20 96GB?

NO — Won't Fit

F0Won't run

○Estimated from fit model

Llama 4 Maverick 17B 128E needs ~257.4 GB but NVIDIA H20 96GB only has 96.0 GB. Try a smaller quantization or lighter model.

Runtime: llama.cppCapacity: No fitBandwidth: HighStack: StandardBottleneck: Memory capacity

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 257.4 GB, exceeds 96.0 GB available

257.4 GB required96.0 GB available

268% VRAM needed

161.4 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

7.2 tok/s

TTFT

26915 ms

Safe context

Memory

257.4 GB / 96.0 GB

Offload

60%

Memory breakdown

Weights244.0 GB

KV Cache2.9 GB

Runtime0.9 GB

Headroom9.6 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsLlama 4 Maverick 17B 128E on NVIDIA H20 96GB
1st promptCold start — includes initialization
>Solve: If a train travels at 60 mph for 2.5 hours, how far does it go?
Let me work through this step by step.

Given information:
 - Speed = 60 mph (miles per hour)
 - Time = 2.5 hours

Formula: Distance = Speed x Time

Calculation:
 Distance = 60 mph x 2.5 h
 = 150 miles

The train travels 150 miles.

Double-check: at 60 mph the train covers 60 miles each hour. In half an hour it covers 30 miles. So in 2.5 hours: 60 + 60 + 30 = 150 miles. Confirmed.
2nd promptWarm — model ready, no init overhead
>What if the train stops for 20 minutes halfway?
The stop doesn't change the total distance covered while moving, but it affects total trip time.

Breakdown:
 - First half: 1.25 hours at 60 mph = 75 miles
 - Stop: 20 minutes (no distance)
 - Second half: 1.25 hours at 60 mph = 75 miles

Total distance: 150 miles (same)
Total time: 2.5 h + 0.33 h = 2.83 hours
Average speed: 150 / 2.83 = 53 mph

The distance stays the same but average speed drops to 53 mph because of the stop.
Estimated: 7.2 tok/s decode · 26.9s TTFT (warm) · 18 tok/s prefill

What limits this setup

Usable VRAM is the main blocker for this model.

Not enough usable memory

The model needs 257.4 GB, but this setup only exposes 96.0 GB of usable VRAM.

Best improvement path

Add more VRAM headroom

The first useful upgrade is more dedicated VRAM so you can fit the model without shrinking context or dropping to a much lower quant.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	F	Too heavy	6.7 tok/s	15707 ms	4K
Coding	F	Too heavy	6.7 tok/s	29068 ms	4K
Agentic Coding	F	Too heavy	6.5 tok/s	43078 ms	4K
Reasoning	F	Too heavy	6.7 tok/s	34353 ms	4K
RAG	F	Too heavy	6.5 tok/s	53847 ms	4K

Quantization options

How Llama 4 Maverick 17B 128E (400B params) fits at each quantization level on NVIDIA H20 96GB (96.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	156.0 GB	Low	F0
Q3_K_S	3	196.0 GB	Low	F0
NVFP4	4

Upgrade options

Hardware that runs Llama 4 Maverick 17B 128E well

AMD Instinct MI350X 288GBBudget pick

288 GB VRAM (+192)8000 GB/s (+4000)

Makes the model fit on the accelerator instead of staying completely out of reach.77.4 tok/s decode

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

~$8,000 MSRP

AMD Instinct MI325X 256GBBest value

256 GB VRAM (+160)6000 GB/s (+2000)

Makes the model fit on the accelerator instead of staying completely out of reach.37.9 tok/s decode