VOOZH about

URL: https://willitrunai.com/can-run/qwen-3-30b-a3b-on-h200-141gb

⇱ Qwen 3 30B A3B on NVIDIA H200 141GB? YES

Can Qwen 3 30B A3B run on NVIDIA H200 141GB?

YES — Runs Great

S88Excellent
○Estimated from fit model

Qwen 3 30B A3B needs ~35.4 GB VRAM. NVIDIA H200 141GB has 141.0 GB. With Q4_K_M quantization, expect ~610 tok/s.

Runtime: OllamaCapacity: RoomyBandwidth: HighStack: BasicBottleneck: Balanced
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.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 35.4 GB, 609.7 tok/s, Runs well
35.4 GB required141.0 GB available
25% VRAM used

Fit status

Runs well

Decode

609.7 tok/s

TTFT

350 ms

Safe context

131K

Memory

35.4 GB / 141.0 GB

Memory breakdown

Weights18.6 GB
KV Cache1.5 GB
Runtime1.2 GB
Headroom14.1 GB

See how fast it feels

See how fast it feelsQwen 3 30B A3B on NVIDIA H200 141GB
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: 609.7 tok/s decode · 350ms TTFT (warm) · 1524 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

No major red flags

This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.

Best improvement path

Performance by workload

WorkloadGradeFitDecodeTTFTContext
ChatSRuns well609.7 tok/s350 ms131K
CodingSRuns well609.7 tok/s350 ms131K
Agentic CodingSRuns well609.7 tok/s462 ms131K
ReasoningSRuns well609.7 tok/s375 ms131K
RAGSRuns well609.7 tok/s577 ms131K

Quantization options

How Qwen 3 30B A3B (30.5B params) fits at each quantization level on NVIDIA H200 141GB (141.0 GB usable).

QuantBitsVRAMQualityFit
Q2_K
2
11.9 GB
LowA79
Q3_K_S
3
14.9 GB
LowA79
NVFP4
4
17.1 GB
MediumA79
Q4_K_M
4
18.6 GB
MediumA79
Q5_K_M
5
22.0 GB
HighA79
Q6_K
6
25.0 GB
HighA80
Q8_0
8
32.6 GB
Very HighA81
F16Best for your GPU
16
62.5 GB
MaximumS85

Get started

Copy-paste commands to run Qwen 3 30B A3B on your machine.

Run

ollama run qwen3:30b-a3b

Your hardware

More models your NVIDIA H200 141GB can run

ModelParamsGradeDecodeCapabilities
👁 Mistral
Devstral 2 123B Instruct		123BS58.4 tok/s
👁 Alibaba
Qwen 3.5 122B A10B		122BS162.1 tok/s
👁 Alibaba
Qwen 3.6 35B A3B		35BS512.4 tok/s
👁 Alibaba
Qwen 3.5 35B A3B		35BS557.2 tok/s
👁 Alibaba
Qwen 3 32B		32BS224.6 tok/s

Frequently asked questions

See all results for NVIDIA H200 141GBSee all hardware for Qwen 3 30B A3B