NVIDIA GeForce RTX 4090 vs NVIDIA L40, AI & Machine Learning Comparison
NVIDIA GeForce RTX 4090 wins 8 of 12 benchmarks, averaging 35.2% faster.
NVIDIA L40's numbers are anchored estimates calibrated against our measured cards, pending first-party measurement. Treat small gaps as ties.
What the numbers say
The gap is widest in LTX-Video (distilled), where NVIDIA GeForce RTX 4090 leads by 80% (4.7 vs 2.61 frames/s); the closest fight is Wan 2.2 5B (720p) (16% apart); VRAM decides part of this one: NVIDIA L40 runs 11 of our 12 AI workloads while the other card runs 8, models that don't fit score zero.
Benchmark results head-to-head
| Benchmark | NVIDIA GeForce RTX 4090 | NVIDIA L40 | Difference |
|---|---|---|---|
| Qwen3 4B tok/s | 260.54 | 211.33 | +23% |
| Llama 3.1 8B tok/s | 171.29 | 135.12 | +27% |
| Qwen2.5-Coder 14B tok/s | 95.12 | 74.27 | +28% |
| Qwen3 32B tok/s | 44.28 | 34.08 | +30% |
| Llama 3.3 70B tok/s | 0 | 0 | n/a |
| Stable Diffusion XL images/min | 16.28 | 11.2 | +45% |
| Z-Image Turbo images/min | 7.425 | 5.625 | +32% |
| FLUX.1 dev images/min | 0 | 2.464 | n/a |
| FLUX.1 Kontext dev images/min | 0 | 1.114 | n/a |
| Qwen-Image-Edit images/min | 0 | 0.48 | n/a |
| LTX-Video (distilled) frames/s | 4.7 | 2.61 | +80% |
| Wan 2.2 5B (720p) frames/s | 0.43 | 0.37 | +16% |
Whole-job comparison
How long each card takes to finish a complete pipeline, not just one model. NVIDIA GeForce RTX 4090 is faster on 4 of 4; NVIDIA L40 on 0.
| Workflow | NVIDIA GeForce RTX 4090 | NVIDIA L40 | Difference | Cost per run |
|---|---|---|---|---|
| 24-frame storyboard | 5 min | 7 min | NVIDIA GeForce RTX 4090 1.40x faster | $0.013 vs $0.080 |
| 60-second AI short film | 7.9 min | 13 min | NVIDIA GeForce RTX 4090 1.63x faster | $0.021 vs $0.149 |
| Full codebase review | 10.5 min | 13.5 min | NVIDIA GeForce RTX 4090 1.28x faster | $0.027 vs $0.155 |
| 10 short social clips | 22.7 min | 28.2 min | NVIDIA GeForce RTX 4090 1.24x faster | $0.059 vs $0.324 |
Renting by the hour, NVIDIA GeForce RTX 4090 finishes 4 of 4 cheaper. The quicker card is not automatically the cheaper way to get the work done.
Cost to rent
| Card | Per hour |
|---|---|
| NVIDIA GeForce RTX 4090 | $0.155 |
| NVIDIA L40 | $0.690 |
NVIDIA GeForce RTX 4090 is 4.45x cheaper per hour. Cheapest on-demand rate we see across RunPod and Vast.
Specifications compared
| NVIDIA GeForce RTX 4090 | NVIDIA L40 | |
|---|---|---|
| VRAM | 24GB | 48GB |
| Architecture | Ada Lovelace | Ada Lovelace |
| Memory bandwidth | 1008 GB/s | 864 GB/s |
| Boost clock | 2,520 MHz | 2,490 MHz |
| TDP | 450 W | 300 W |
| Launch MSRP | $1,599 | $7,000 |
| Release | 2022-10-12 | 2022-10-13 |
FAQ
- Which is better for ai & machine learning: NVIDIA GeForce RTX 4090 or NVIDIA L40?
- NVIDIA GeForce RTX 4090 performs better for ai & machine learning, winning 8 of 12 benchmarks in our suite with an average 35.2% advantage.
- What are the main hardware differences between NVIDIA GeForce RTX 4090 and NVIDIA L40?
- NVIDIA GeForce RTX 4090 has 24GB VRAM and a 450W TDP, while NVIDIA L40 has 48GB VRAM and a 300W TDP.
- Does VRAM matter more than speed between NVIDIA GeForce RTX 4090 and NVIDIA L40?
- For AI, yes, NVIDIA L40 fits 4 more of our 12 workloads than NVIDIA GeForce RTX 4090. A model that exceeds VRAM doesn't run slower, it doesn't run at all, so the card that fits the model wins that workload outright.
- Where is the biggest performance difference between NVIDIA GeForce RTX 4090 and NVIDIA L40?
- LTX-Video (distilled): NVIDIA GeForce RTX 4090 leads by roughly 80% (4.7 vs 2.61 frames/s) in our testing.
NVIDIA GeForce RTX 4090 full review · NVIDIA L40 full review · All AI & Machine Learning rankings