Wan 2.2 T2V A14B · 3 GPUs measured first-party · text-to-video · Updated October 2026
What GPU Do You Need for Wan 2.2 T2V A14B?
Wan 2.2 T2V A14B on 3 GPUs, measured first-party: NVIDIA H200 leads at 0.35 frames/s, A100 80GB SXM4 trails at 0.17 frames/s, and it peaked at 75GB of VRAM.
Benchmarked weights: Wan-AI/Wan2.2-T2V-A14B-Diffusers
NVIDIA H200
0.35 frames/s on Wan 2.2 T2V A14B, the ceiling. Measured on our bench. 141GB of VRAM, $31,000 at launch.
- 0.35 frames/s on Wan 2.2 T2V A14B
- 141GB, clears the Wan 2.2 T2V A14B floor
- Rentable by the hour rather than bought
- 700W board rating
- Datacenter or workstation hardware, not a retail purchase
What GPU Do You Need for Wan 2.2 T2V A14B?, frames/s by GPU
Efficiency: frames/s per 100W drawn
Power is the average pulled during the run, sampled at 1Hz. The fastest card is often not the one here, and for anything left running this is the number that shows up on the bill.
Value: frames/s per $1,000 of MSRP
Launch price, not street price, so it ages. A speed leaderboard always crowns the most expensive card; this is the counterweight.
Wan 2.2 T2V A14B. Measured text-to-video speed by GPU
| GPU | frames/s | Peak VRAM | Avg power |
|---|---|---|---|
| NVIDIA H200 | 0.35 | 71.1GB | 688.2 W |
| NVIDIA H100 80GB HBM3 | 0.34 | 71.0GB | 689.4 W |
| NVIDIA A100 80GB SXM4 | 0.17 | 70.8GB | 394.9 W |
What the numbers show. Across 3 GPUs measured on our own bench, H200 is fastest at 0.35 frames/s. The slowest, A100 80GB SXM4, manages 0.17, so the spread is 2.1x from top to bottom.
About Wan 2.2 T2V A14B. Wan 2.2 T2V A14B: from Wan-AI, 14B parameters, on Hugging Face since July 2025, Apache 2.0 licence. 378,526 downloads in the last 30 days and 9 community quantizations.
How it compares. H200: Wan 2.2 T2V A14B 0.35 frames/s, Wan 2.1 T2V 14B 0.36 (14B), Mochi 1 Preview 0.4 (10B), LTX-Video (distilled) 17.87 (2B), Wan 2.2 5B (720p) 1.41. All 4 beat Wan 2.2 T2V A14B here.
Cost on a rented GPU. 1 minute of 24fps video of Wan 2.2 T2V A14B: $2.23 on a A100 80GB SXM4 ($0.95/hr, 2.4 hours), $4.11 on a H200 ($3.59/hr, 69 min, 1.8x the cost).
Wan 2.2 T2V A14B: cost per 1 minute of 24fps video on rented GPUs
| GPU | Cheapest rate | Speed (frames/s) | Cost per 1 minute of 24fps video |
|---|---|---|---|
| NVIDIA A100 80GB SXM4 | $0.95/hr | 0.17 | $2.23 |
| NVIDIA H100 80GB HBM3 | $2.14/hr | 0.34 | $2.55 |
| NVIDIA H200 | $3.59/hr | 0.35 | $4.11 |
Cheapest hourly rate we track on RunPod and Vast.ai, divided by the measured speed. Startup time and storage are extra.
Speed tiers for Wan 2.2 T2V A14B. under 1 frames/s: 3 (H200, H100 80GB HBM3, A100 80GB SXM4). 10 frames/s turns out a minute of 24fps video in under 2.5 minutes.
VRAM for Wan 2.2 T2V A14B. Measured peak 70.8GB, so 80GB is the smallest common card size; smallest card it ran on: H100 80GB HBM3 (80GB).
Power on Wan 2.2 T2V A14B. Most efficient: H200, 688W, 0.79 kWh per 1 minute of 24fps video. Hungriest: H100 80GB HBM3, 689W, 0.82 kWh. At $0.15/kWh: $0.12 per 1 minute of 24fps video.
Our verdict
Fastest on Wan 2.2 T2V A14B: NVIDIA H200, 0.35 frames/s. Cheapest to rent per job: NVIDIA A100 80GB SXM4, $2.23 per 1 minute of 24fps video.
FAQ
- What GPU do I need to run Wan 2.2 T2V A14B?
- About 74GB. Smallest card that ran it: NVIDIA H100 80GB HBM3 (80GB).
- How much does it cost to run Wan 2.2 T2V A14B in the cloud?
- $2.23 per 1 minute of 24fps video on a NVIDIA A100 80GB SXM4 at $0.95/hr, cheapest of 3 rentable cards we measured.
- Can I run Wan 2.2 T2V A14B on a 12GB, 16GB or 24GB card?
- It used 70.8GB at the precision we tested. 12GB: no; 16GB: no; 24GB: no.
- Is the H100 80GB HBM3 or the A100 80GB SXM4 faster for Wan 2.2 T2V A14B?
- The H100 80GB HBM3: 0.34 vs 0.17 frames/s, 97% faster on our bench.
How we test
Wan 2.2 T2V A14B in diffusers, bf16, one warmup clip and three timed clips from the same prompt, measured as frames of finished video per second, with CPU offload only on cards below the model's full-GPU size. Text-to-video is compute-bound and memory-hungry; cards that need CPU offload lose far more speed than their raw compute suggests.