GPUs with 20–∞ GB VRAM for Running LLMs
Browse 51 GPUs with 20–∞ GB VRAM compatible with running LLM models locally. Compare VRAM, memory bandwidth, and AI performance. Top picks and budget guide
51 GPUs
| GPU | Price |
|---|---|
| RTX 409024 GBGDDR6X · 1,008 GB/s · 450 WNVIDIA · Ada Lovelace · 2022 | about $1,599 |
| RTX 509032 GBGDDR7 · 1,792 GB/s · 575 WNVIDIA · Blackwell · 2025 | about $5,376 |
| RTX 309024 GBGDDR6X · 936 GB/s · 350 WNVIDIA · Ampere · 2020 | about $1,499 |
| RTX PRO 6000 BlackwellPro96 GBGDDR7 · 1,792 GB/s · 600 WNVIDIA · Blackwell · 2025 | about $8,565 |
| RX 7900 XTX24 GBGDDR6 · 960 GB/s · 355 WAMD · RDNA 3 · 2022 | about $1,130 |
| H200 SXMPro141 GBHBM3e · 4,800 GB/s · 700 WNVIDIA · Hopper · 2024 | – |
| Radeon AI PRO R9700Pro32 GBGDDR6 · 640 GB/s · 300 WAMD · RDNA 4 · 2025 | about $1,299 |
| B200Pro192 GBHBM3e · 8,000 GB/s · 1000 WNVIDIA · Blackwell · 2024 | – |
| RTX 3090 Ti24 GBGDDR6X · 1,008 GB/s · 450 WNVIDIA · Ampere · 2022 | about $1,999 |
| H100 SXMPro80 GBHBM3 · 3,352 GB/s · 700 WNVIDIA · Hopper · 2022 | – |
| Tesla P40Pro24 GBGDDR5 · 346 GB/s · 250 WNVIDIA · Pascal · 2016 | – |
| Arc Pro B60Pro24 GBGDDR6 · 456 GB/s · 200 WIntel · Xe2 · 2025 | about $599 |
| Arc Pro B70Pro32 GBGDDR6 · 608 GB/s · 230 WIntel · Xe2 · 2026 | about $949 |
| H100 PCIePro80 GBHBM2e · 2,039 GB/s · 350 WNVIDIA · Hopper · 2022 | – |
| Arc Pro B60 Dual 48GBPro48 GBGDDR6 · 456 GB/s · 400 WIntel · Xe2 · 2025 | about $1,200 |
| RTX 5090 Laptop GPULaptop24 GBGDDR7 · 896 GB/s · 150 WNVIDIA · Blackwell · 2025 | – |
| Arc Pro B65Pro32 GBGDDR6 · 608 GB/s · 200 WIntel · Xe2 · 2026 | about $910 |
| RX 7900 XT20 GBGDDR6 · 800 GB/s · 315 WAMD · RDNA 3 · 2022 | about $899 |
| A100 80GB SXMPro80 GBHBM2e · 2,039 GB/s · 400 WNVIDIA · Ampere · 2020 | – |
| RTX PRO 5000 BlackwellPro72 GBGDDR7 · 1,344 GB/s · 300 WNVIDIA · Blackwell · 2025 | about $4,500 |
| RTX PRO 6000 Blackwell Max-QPro96 GBGDDR7 · 1,792 GB/s · 300 WNVIDIA · Blackwell · 2025 | about $8,565 |
| A100 40GB PCIePro40 GBHBM2e · 1,555 GB/s · 250 WNVIDIA · Ampere · 2020 | – |
| H200 NVLPro141 GBHBM3e · 4,800 GB/s · 600 WNVIDIA · Hopper · 2024 | – |
| Tesla M40 24GBPro24 GBGDDR5 · 288 GB/s · 250 WNVIDIA · Maxwell · 2015 | – |
| RTX PRO 4000 BlackwellPro24 GBGDDR7 · 672 GB/s · 145 WNVIDIA · Blackwell · 2025 | about $1,500 |
Which GPU Do You Need for AI?
The amount of VRAM is the most important specification for running LLMs locally. Most 7B parameter models require 4–8 GB of VRAM at common quantization levels, while 70B models need 24–48 GB. Memory bandwidth determines how fast the model generates tokens — faster bandwidth means faster responses.