The best GPU for running Qwen3-VL 30B-A3B Instruct (MoE) (2026)
Ranked from live rental prices and computed VRAM fit, not opinion. Every card is judged at Q4_K_M — the quantisation most people actually run — at 8k context, so it's a fair comparison. A card that can't fit Qwen3-VL 30B-A3B Instruct (MoE)at that quality isn't listed, because it could only run a crushed version.
Best value
NVIDIA RTX A5000
Best throughput per dollar among cards that fit.
Cheapest that runs it
NVIDIA RTX A5000
Lowest hourly rental that fits it — $0.077/hr, Q4_K_M.
No compromise
—
Most headroom and highest throughput.
Every card that runs it, ranked
Best quantisation each card fits at 8k context, with the cheapest live rental and a bandwidth-derived throughput estimate. Sorted by tokens/sec per dollar.
| GPU | VRAM | VRAM used | ~tok/s | Cheapest rental | tok/s per $ |
|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | 96 GB | 19 GB20% | — | $0.467 | — |
| NVIDIA RTX PRO 6000 Blackwell Workstation Edition | 96 GB | 19 GB20% | — | $0.533 | — |
| NVIDIA RTX 6000 Ada Generation | 48 GB | 19 GB40% | — | $0.200 | — |
| AMD Radeon PRO W7900 | 48 GB | 19 GB40% | — | — | — |
| NVIDIA RTX A6000 | 48 GB | 19 GB40% | — | $0.107 | — |
| NVIDIA GeForce RTX 5090 | 32 GB | 19 GB60% | — | $0.267 | — |
| NVIDIA RTX 5000 Ada Generation | 32 GB | 19 GB60% | — | $0.267 | — |
| NVIDIA GeForce RTX 4090 | 24 GB | 19 GB79% | — | $0.170 | — |
| AMD Radeon RX 7900 XTX | 24 GB | 19 GB79% | — | — | — |
| NVIDIA GeForce RTX 3090 Ti | 24 GB | 19 GB79% | — | $0.111 | — |
| NVIDIA RTX A5000 | 24 GB | 19 GB79% | — | $0.077 | — |
| NVIDIA GeForce RTX 3090 | 24 GB | 19 GB79% | — | $0.080 | — |
How this ranking is made — and its limit
Fit and throughput are computed from Qwen3-VL 30B-A3B Instruct (MoE)'s own configuration with the engine behind our cost-to-run page (bits-per-weight validated to 0.7% median error). Rental prices are the cheapest live figure across the providers we track, last updated Sat, 07 Nov 2026 02:20:13 GMT.
The value column ranks renting, because that's what we can price precisely. If you're buyinga card, the fit and throughput columns are exactly what you need — but we don't publish a purchase-price value ranking, because we don't have verified street prices and won't invent them. For the buy-vs-rent decision itself, see the guides below.
Tokens/sec is a bandwidth-derived estimate, not a benchmark — we don't run our own hardware tests (editorial policy).
Check it yourself
See exactly what fits, at any context length
Running Qwen3-VL 30B-A3B Instruct (MoE): common questions
How much VRAM do you need to run Qwen3-VL 30B-A3B Instruct (MoE)?
Can an RTX 4090 run Qwen3-VL 30B-A3B Instruct (MoE)?
Can an RTX 3090 run Qwen3-VL 30B-A3B Instruct (MoE)?
Can an RTX 4060 Ti 16GB run Qwen3-VL 30B-A3B Instruct (MoE)?
What's the cheapest way to run Qwen3-VL 30B-A3B Instruct (MoE)?
Before you buy
How much VRAM do you need?
About 0.57 GB per billion parameters at 4-bit, plus room for context: the rule behind the numbers on this page, with worked examples.
The best GPU for local AI right now
The card with enough VRAM for your model at the lowest price, budget by budget.
Rent a cloud GPU or buy one?
Rent, unless you'll pass about 2,000 hours of use: the break-even and the costs people forget on both sides.
What quantization means
Why a 4-bit copy of a model fits on a normal PC, and what you give up for it.