Can Mixtral 8x22B run on NVIDIA H100 80GB?

Q: What should I upgrade first if Mixtral 8x22B feels slow on NVIDIA H100 80GB?

Remove offload with more accelerator memory. Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

YES — With NVFP4

B56Good

Estimated from fit model

Mixtral 8x22B needs ~91.3 GB VRAM. NVIDIA H100 80GB has 80.0 GB. With NVFP4 quantization, expect ~53 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: HighStack: StandardBottleneck: Host offload

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.

Mixtral 8x22B at Q4_K_M needs 98.3 GB — too much for NVIDIA H100 80GB (80.0 GB). Runs at NVFP4 (91.3 GB) with medium quality. 3 quantization levels fit.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 98.3 GB, exceeds 80.0 GB available

98.3 GB required80.0 GB available

123% VRAM needed

18.3 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

41.3 tok/s

TTFT

4692 ms

Safe context

Memory

98.3 GB / 80.0 GB

Offload

20%

Memory breakdown

Weights86.0 GB

KV Cache3.4 GB

Runtime0.9 GB

Headroom8.0 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsMixtral 8x22B on NVIDIA H100 80GB
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: 41.3 tok/s decode · 4.7s TTFT (warm) · 103 tok/s prefill

What limits this setup

It fits through host-memory offload, and offload is the main reason performance drops.

CPU or host-memory offload is active

About 10% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.

Very little memory headroom

You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.

Best improvement path

Remove offload with more accelerator memory

Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Buy headroom, not only minimum fit

A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

Increase host RAM if you keep offloading

This setup may need roughly 9.8 GB of extra host RAM just for the offloaded portion, before OS and other tools.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	F	Too heavy	42.5 tok/s	2486 ms	4K
Coding	F	Too heavy	38.1 tok/s	5083 ms	4K
Agentic Coding	F	Too heavy	39.0 tok/s	7220 ms	4K
Reasoning	F	Too heavy	41.3 tok/s	5545 ms	4K
RAG	F	Too heavy	39.0 tok/s	9025 ms	4K

Quantization options

How Mixtral 8x22B (141B params) fits at each quantization level on NVIDIA H100 80GB (80.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	55.0 GB	Low	B61
Q3_K_S	3	69.1 GB	Low	F0

Get started

Copy-paste commands to run Mixtral 8x22B on your machine.

Run

ollama run mixtral:8x22b

Upgrade options

Hardware that runs Mixtral 8x22B well

RTX PRO 6000 Blackwell Workstation Edition 96GBBudget pick

96 GB VRAM (+16)

Makes the model fit on the accelerator instead of staying completely out of reach.25.6 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.

~$9,999 MSRP

RTX PRO 6000 Blackwell Server Edition 96GBBest value

96 GB VRAM (+16)

Makes the model fit on the accelerator instead of staying completely out of reach.22.8 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.

~$9,999 MSRP

NVIDIA H20 96GBNVIDIA upgrade

96 GB VRAM (+16)4000 GB/s (+650)

Makes the model fit on the accelerator instead of staying completely out of reach.62.5 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.

~$12,000 MSRP

Frequently asked questions

Can NVIDIA H100 80GB run Mixtral 8x22B?

Yes, NVIDIA H100 80GB can run Mixtral 8x22B at NVFP4 quantization (Very compromised (needs ~9.8 GB host RAM)). The recommended Q4_K_M requires 98.3 GB which exceeds available memory, but at NVFP4 it needs only 91.3 GB. Expected decode speed: 53.4 tok/s.

How much VRAM does Mixtral 8x22B need?

Mixtral 8x22B (141B parameters) requires approximately 98.3 GB at Q4_K_M quantization. On NVIDIA H100 80GB, it fits at NVFP4 using 91.3 GB.

What is the best quantization for Mixtral 8x22B?

The recommended quantization is Q4_K_M, but on NVIDIA H100 80GB the best fitting quantization is NVFP4, which uses 91.3 GB.

What speed will Mixtral 8x22B run at on NVIDIA H100 80GB?

On NVIDIA H100 80GB, Mixtral 8x22B achieves approximately 53.4 tokens per second decode speed with a time-to-first-token of 3628ms using NVFP4 quantization.

Can NVIDIA H100 80GB run Mixtral 8x22B for coding?

For coding workloads, Mixtral 8x22B on NVIDIA H100 80GB receives a F grade with 38.1 tok/s and 4K context.

What context window can Mixtral 8x22B use on NVIDIA H100 80GB?

On NVIDIA H100 80GB, Mixtral 8x22B can safely use up to 4K tokens of context at NVFP4 quantization. The model's official context limit is 66K, but available memory constrains the safe maximum.

See all results for NVIDIA H100 80GB See all hardware for Mixtral 8x22B

Embed this result▼

Paste this snippet into any page to show a live fit card.

<iframe src="https://willitrunai.com/embed/mixtral-8x22b-on-h100-80gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: