Will It Run AI

Can Hermes 4.3 36B run on NVIDIA A30 24GB?

YES — With NVFP4

D39Poor
Estimated from fit model

Hermes 4.3 36B needs ~28.0 GB VRAM. NVIDIA A30 24GB has 24.0 GB. With NVFP4 quantization, expect ~21 tok/s.

Runtime: OllamaCapacity: OffloadBandwidth: HighStack: BasicBottleneck: Host offload
Share:

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.

Hermes 4.3 36B at Q4_K_M needs 29.8 GB — too much for NVIDIA A30 24GB (24.0 GB). Runs at NVFP4 (28.0 GB) with medium quality. 3 quantization levels fit.
Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 29.8 GB, exceeds 24.0 GB available
29.8 GB required24.0 GB available
124% VRAM needed

5.8 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

15.8 tok/s

TTFT

12267 ms

Safe context

4K

Memory

29.8 GB / 24.0 GB

Offload

20%

Memory breakdown

Weights22.0 GB
KV Cache4.2 GB
Runtime1.2 GB
Headroom2.4 GB

See how fast it feels

With memory offload — actual speed may be lower
See how fast it feelsHermes 4.3 36B on NVIDIA A30 24GB
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: 15.8 tok/s decode · 12.3s TTFT (warm) · 40 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 2.9 GB of extra host RAM just for the offloaded portion, before OS and other tools.

Performance by workload

WorkloadGradeFitDecodeTTFTContext
ChatDVery compromised (needs ~2.9 GB host RAM)18.4 tok/s5732 ms4K
CodingFToo heavy15.8 tok/s12267 ms4K
Agentic CodingFToo heavy11.9 tok/s23582 ms4K
ReasoningFToo heavy15.8 tok/s14497 ms4K
RAGFToo heavy11.9 tok/s29477 ms4K

Quantization options

How Hermes 4.3 36B (36B params) fits at each quantization level on NVIDIA A30 24GB (24.0 GB usable).

QuantBitsVRAMQualityFit
Q2_K
2
14.0 GB
LowC50
Q3_K_SBest for your GPU
3
17.6 GB
LowC49
NVFP4
4
20.2 GB
MediumF0
Q4_K_M
4
22.0 GB
MediumF0
Q5_K_M
5
25.9 GB
HighF0
Q6_K
6
29.5 GB
HighF0
Q8_0
8
38.5 GB
Very HighF0
F16
16
73.8 GB
MaximumF0

Get started

Copy-paste commands to run Hermes 4.3 36B on your machine.

Run

lms load hf-nousresearch--hermes-4-3-36b-gguf && lms server start

Opciones de mejora

Hardware que ejecuta bien Hermes 4.3 36B

NVIDIARTX 5090 32GBOpción económica
32 GB VRAM (+8)1792 GB/s (+859)
C
Hace que el modelo quepa en el acelerador en lugar de seguir fuera de alcance.54.7 tok/s decodificación

Hace que el modelo quepa en el acelerador en lugar de seguir fuera de alcance.

Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.

~$1,999 MSRP

NVIDIARTX PRO 4500 Blackwell 32GBMejor relación calidad-precio
32 GB VRAM (+8)
C
Hace que el modelo quepa en el acelerador en lugar de seguir fuera de alcance.34.3 tok/s decodificación

Hace que el modelo quepa en el acelerador en lugar de seguir fuera de alcance.

Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.

~$2,499 MSRP

NVIDIARTX 5000 Ada 32GBMejora NVIDIA
32 GB VRAM (+8)
C
Hace que el modelo quepa en el acelerador en lugar de seguir fuera de alcance.21 tok/s decodificación

Hace que el modelo quepa en el acelerador en lugar de seguir fuera de alcance.

Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.

~$4,000 MSRP

Frequently asked questions

Can NVIDIA A30 24GB run Hermes 4.3 36B?

Yes, NVIDIA A30 24GB can run Hermes 4.3 36B at NVFP4 quantization (Very compromised (needs ~2.9 GB host RAM)). The recommended Q4_K_M requires 29.8 GB which exceeds available memory, but at NVFP4 it needs only 28.0 GB. Expected decode speed: 20.6 tok/s.

How much VRAM does Hermes 4.3 36B need?

Hermes 4.3 36B (36B parameters) requires approximately 29.8 GB at Q4_K_M quantization. On NVIDIA A30 24GB, it fits at NVFP4 using 28.0 GB.

What is the best quantization for Hermes 4.3 36B?

The recommended quantization is Q4_K_M, but on NVIDIA A30 24GB the best fitting quantization is NVFP4, which uses 28.0 GB.

What speed will Hermes 4.3 36B run at on NVIDIA A30 24GB?

On NVIDIA A30 24GB, Hermes 4.3 36B achieves approximately 20.6 tokens per second decode speed with a time-to-first-token of 9406ms using NVFP4 quantization.

Can NVIDIA A30 24GB run Hermes 4.3 36B for coding?

For coding workloads, Hermes 4.3 36B on NVIDIA A30 24GB receives a F grade with 15.8 tok/s and 4K context.

What context window can Hermes 4.3 36B use on NVIDIA A30 24GB?

On NVIDIA A30 24GB, Hermes 4.3 36B can safely use up to 4K tokens of context at NVFP4 quantization. The model's official context limit is —, but available memory constrains the safe maximum.

What should I upgrade first if Hermes 4.3 36B feels slow on NVIDIA A30 24GB?

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.

See all results for NVIDIA A30 24GBSee all hardware for Hermes 4.3 36B
Embed this result▼

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

<iframe src="https://willitrunai.com/embed/hf-nousresearch--hermes-4-3-36b-gguf-on-a30-24gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: