24h drops
GeForce RTX 4060 Ti 16GB$1,327-16.4%- GeForce RTX 3060 12GB$437-15.7%
- GeForce RTX 3090$2,497-9.2%
- RTX A6000$8,778-7.9%
- GeForce RTX 5060 Ti 16GB$923-2.9%
24h rises
- RTX PRO 4000$4,290+43.7%
- GeForce RTX 5090$9,360+20.4%
- Radeon RX 7900 XT$1,393+14.7%
- GeForce RTX 4090$5,009+4.6%
- GeForce RTX 4080$2,376+2.4%
Most listings
- GeForce RTX 5080$1,99316 listings
- GeForce RTX 5060 Ti 16GB$92314 listings
- Radeon RX 9070 XT$89514 listings
- GeForce RTX 4080$2,37613 listings
- GeForce RTX 5070 Ti$1,45713 listings
Latest release
DeepSeek V4.1 Flash763BQ4444.7 GB
Hy4 preview770BQ4467.3 GB
GLM 5.3744BQ4467.3 GB- GLM 5.3 Flash320BQ4199.7 GB
Qwen3.8 Flash Next176BQ4111.3 GB
Local LLM GPU Ranking
Amazon.jp
| Model | 30d trend | Ecosystem | |||||||
|---|---|---|---|---|---|---|---|---|---|
| H100 PCIe | No price in this market | — | 80 | 2,000 | 756 | 1,513 | 350 | CUDA | |
| RTX PRO 6000 | No price in this market | — | 96 | 1,792 | 500 | 1,000 | 600 | CUDA | |
| GeForce RTX 5090 | $9,3604 listings | +20.4% | 32 | 1,792 | 419 | 838 | 575 | CUDA | |
| RTX PRO 6000D | No price in this market | — | 84 | 1,398 | 385.2 | 770.5 | 600 | CUDA | |
| L40S | No price in this market | — | 48 | 864 | 362.1 | 733 | 350 | CUDA | |
| RTX 6000 Ada | No price in this market | — | 48 | 960 | 364.3 | 728.5 | 300 | CUDA | |
| GeForce RTX 4090 | $5,0098 listings | +4.6% | 24 | 1,008 | 330.3 | 660.6 | 450 | CUDA | |
| GeForce RTX 4090 48GB | No price in this market | — | 48 | 1,008 | 330.3 | 660.6 | 450 | CUDA | |
| A100 40GB PCIe | No price in this market | — | 40 | 1,555 | 312 | 624 | 250 | CUDA | |
| A100 80GB PCIe | No price in this market | — | 80 | 1,935 | 312 | 624 | 300 | CUDA | |
| GeForce RTX 5090 D | No price in this market | — | 32 | 1,792 | 296.9 | 593.8 | 575 | CUDA | |
| GeForce RTX 5090 D V2 | No price in this market | — | 24 | 1,344 | 296.9 | 593.8 | 575 | CUDA | |
| GeForce RTX 4090D 48GB | No price in this market | — | 48 | 1,008 | 294.3 | 588.5 | 425 | CUDA | |
| RTX PRO 5000 48GB | No price in this market | — | 48 | 1,344 | 258 | 516 | 300 | CUDA | |
| RTX PRO 5000 72GB | No price in this market | — | 72 | 1,344 | 258 | 516 | 300 | CUDA | |
| GeForce RTX 5080 | $1,99316 listings | +0.4% | 16 | 960 | 225.1 | 450.2 | 360 | CUDA | |
| GeForce RTX 4080 SUPER 32GB | No price in this market | — | 32 | 736 | 208.9 | 417.8 | 320 | CUDA | |
| RTX PRO 4500 | No price in this market | — | 32 | 896 | 202.1 | 404.3 | 200 | CUDA | |
| CMP 170HX 40GB | No price in this market | — | 40 | 1,560 | 202 | 404 | 250 | CUDA | |
| CMP 170HX 64GB | No price in this market | — | 64 | 1,493 | 202 | 404 | 250 | CUDA | |
| GeForce RTX 4080 | $2,37613 listings | +2.4% | 16 | 717 | 195 | 389.9 | 320 | CUDA | |
| Radeon RX 9070 XT | $89514 listings | +0.4% | 16 | 640 | 195 | 389 | 304 | ROCm | |
| Radeon AI PRO R9700 | $2,2817 listings | 0.0% | 32 | 640 | 191 | 383 | 300 | ROCm | |
 Arc Pro B70 | $2,1893 listings | 0.0% | 32 | 608 | 183.5 | 367 | 230 | oneAPI | |
| GeForce RTX 5070 Ti | $1,45713 listings | -0.1% | 16 | 896 | 175.8 | 351.5 | 300 | CUDA | |
| RTX PRO 4000 | $4,2905 listings | +43.7% | 24 | 672 | 161.3 | 322.5 | 145 | CUDA | |
| RTX A6000 | $8,7784 listings | -7.9% | 48 | 768 | 154.9 | 309.7 | 300 | CUDA | |
| GeForce RTX 3090 | $2,4979 listings | -9.2% | 24 | 936 | 142.3 | 284.7 | 350 | CUDA | |
| DGX Spark 128GB | No price in this market | — | 128 | 273 | 125 | 250 | 240 | CUDA | |
| GeForce RTX 5070 | $1,04513 listings | +1.1% | 12 | 672 | 123.5 | 247 | 250 | CUDA | |
| GeForce RTX 2080 Ti 22GB | No price in this market | — | 22 | 616 | 107.6 | 215.2 | 250 | CUDA | |
| Arc Pro B60 | $1,1147 listings | -0.9% | 24 | 456 | 98.5 | 197 | 200 | oneAPI | |
| Arc Pro B65 | No price in this market | — | 32 | 608 | 98.5 | 197 | 200 | oneAPI | |
| GeForce RTX 5060 Ti 16GB | $92314 listings | -2.9% | 16 | 448 | 94.9 | 189.8 | 180 | CUDA | |
| Instinct MI210 | No price in this market | — | 64 | 1,638 | 181 | 181 | 300 | ROCm | |
| GeForce RTX 4060 Ti 16GB | $1,3273 listings | -16.4% | 16 | 288 | 88.3 | 176.5 | 165 | CUDA | |
| Tesla T10 | No price in this market | — | 16 | 403 | 80 | 160 | 150 | CUDA | |
| Radeon PRO W7900 | No price in this market | — | 48 | 864 | 123 | 123 | 295 | ROCm | |
| Radeon RX 7900 XTX | $2,2299 listings | 0.0% | 24 | 960 | 123 | 123 | 355 | ROCm | |
| Radeon RX 7900 XT | $1,39313 listings | +14.7% | 20 | 800 | 103 | 103 | 315 | ROCm | |
| GeForce RTX 3060 12GB | $4377 listings | -15.7% | 12 | 360 | 51 | 101.9 | 170 | CUDA | |
| Radeon RX 6950 XT | $1,5764 listings | 0.0% | 16 | 576 | 47.3 | 94.6 | 335 | ROCm | |
| Instinct MI100 | No price in this market | — | 32 | 1,229 | 92.3 | 92.3 | 300 | ROCm | |
| Radeon PRO W7800 | No price in this market | — | 32 | 576 | 90.4 | 90.4 | 260 | ROCm | |
| Radeon RX 7800 XT | $1,10711 listings | +2.2% | 16 | 624 | 74.7 | 74.7 | 263 | ROCm | |
| Ryzen AI Max+ 395 128GB | $4,1739 listings | +1.6% | 128 | 256 | 59.4 | 59.4 | 120 | ROCm | |
| Tesla V100 16GB | $7083 listings | 0.0% | 16 | 900 | 112 | 56.1 | 250 | CUDA | |
| Tesla V100 32GB | No price in this market | — | 32 | 900 | 112 | 56.1 | 250 | CUDA | |
| Instinct MI50 32GB | No price in this market | — | 32 | 1,024 | 26.5 | 53 | 300 | ROCm | |
| Tesla P40 | No price in this market | — | 24 | 347 | — | 47 | 250 | CUDA | |
| Mac Studio M3 Ultra 512GB | No price in this market | — | 512 | 819 | — | — | 480 | Metal | |
| Mac Studio M4 Max 128GB | No price in this market | — | 128 | 546 | — | — | 480 | Metal | |
| Mac Studio M5 Max 128GB | No price in this market | — | 128 | 614 | — | — | — | Metal | |
| MacBook Pro M4 Max 128GB | No price in this market | — | 128 | 546 | — | — | 140 | Metal | |
| MacBook Pro M5 Max 128GB | No price in this market | — | 128 | 614 | — | — | 140 | Metal | |
| RTX PRO 5500 | No price in this market | — | 84 | 1,398 | — | — | 600 | CUDA |
No GPUs match “”.
Log in to add GPUs to your watchlist and see it on any device. Log in
No GPUs in your watchlist yet. Click the ☆ before a model name to add one.
DDR4 RAM prices
Amazon.jp
| Spec | Price per stick | Price per GB | Max channels | Max bandwidth | Max memory |
|---|---|---|---|---|---|
| 16 GB | $12614 listings | $7.9 | 2 | 51.2 GB/s | 64 GB4 sticks × 16 GB |
| 32 GB | $2894 listings | $9 | 2 | 51.2 GB/s | 128 GB4 sticks × 32 GB |
| 16 GB | $2147 listings | $13.3 | 8 | 170.6 GB/s | 128 GB8 sticks × 16 GB |
| 32 GB | $37212 listings | $11.6 | 8 | 170.6 GB/s | 256 GB8 sticks × 32 GB |
DDR5 RAM prices
Amazon.jp
| Spec | Price per stick | Price per GB | Max channels | Max bandwidth | Max memory |
|---|---|---|---|---|---|
| 16 GB | $24215 listings | $15.1 | 2 | 96 GB/s | 64 GB4 sticks × 16 GB |
| 32 GB | $4944 listings | $15.4 | 2 | 96 GB/s | 128 GB4 sticks × 32 GB |
| 32 GB | $2,05716 listings | $64 | 12 | 537.6 GB/s | 384 GB12 sticks × 32 GB |
| 64 GB | $3,7166 listings | $58 | 12 | 537.6 GB/s | 768 GB12 sticks × 64 GB |
Choose hardware for local LLM deployment
Find the right GPU for local LLMs. Compare VRAM, GPU prices and model memory requirements to choose hardware that fits your models and budget.
What is VRAM, and how much do you need?
VRAM is the fast memory on a graphics card. Local LLM inference needs space for model weights, the context (KV) cache and runtime buffers. Quantization can reduce weight memory, while longer prompts and more simultaneous users need additional memory. A model fitting in VRAM does not guarantee a particular speed.
Plan a local deployment
Choose a model and quantization, set the context length, and check both GPU memory and system RAM. CPU or expert offloading can make a large model fit with less VRAM, but it depends on RAM bandwidth and can be much slower. Check CUDA, ROCm or Metal support before buying hardware.
Read the rankings and price data
Compare VRAM capacity first, then bandwidth, power and software support. Our prices are median asking prices, not completed sales; sample counts, observation dates and stale-price labels help you judge their usefulness. INT8 throughput alone is not an LLM benchmark.
