Home / Multi-GPU
llama.cpp · homogeneous 2–4 GPUs · not a PC builder
Can this model run across multiple GPUs?
Phase 8 answers memory feasibility for identical cards under llama.cpp. Default split is layer (pipeline parallel: each GPU owns a slice of layers; KV stays with those layers). This is not 2 × VRAM = pool, and it is not a complete PC build.
Cheapest current GPU-only multi-GPU configuration in RigForAI
Homogeneous llama.cpp layer configs that fully fit, among cards with a fresh matched Amazon offer. Cost = GPU count × current per-card price. This is not a complete build and does not prove that N units are in stock.
| # | Configuration | Per-GPU peak | GPU-only cost | Winning SKU |
|---|---|---|---|---|
| 1 | 4× NVIDIA Quadro P5000 16 GB | 12.9 GB | $1323.96 4 × $330.99 Amazon | HP NVIDIA Quadro P5000 (16GB) GDDR5X ASIN B01N4GJ0HL |
| 2 | 3× NVIDIA Quadro P6000 24 GB | 16.9 GB | $1940.88 3 × $646.96 Amazon | HP NVIDIA Quadro P6000 (24GB) Graphics Card ASIN B01NALQ0PB |
| 3 | 4× NVIDIA RTX 5000 16 GB | 12.9 GB | $2411.96 4 × $602.99 Amazon | HP 5JH81AA graphics card NVIDIA Quadro RTX 5000 16 GB GDDR6 ASIN B07PJSNX3F |
| 4 | 4× AMD Radeon RX 6800 16 GB | 12.9 GB | $2519.96 4 × $629.99 Amazon | GIGABYTE AORUS Radeon RX 6800 MASTER 16G AMD 16 GB GDDR6 ASIN B08SGTGTZD |
| 5 | 4× NVIDIA RTX 4060 Ti 16 GB | 12.9 GB | $2519.96 4 × $629.99 Amazon | MSI VENTUS GEFORCE RTX 4060 TI 2X BLACK 16G OC graphics card NVIDIA 16 GB GDDR6 ASIN B0CBK7BRL9 |
| 6 | 4× NVIDIA Quadro P6000 24 GB | 12.9 GB | $2587.84 4 × $646.96 Amazon | HP NVIDIA Quadro P6000 (24GB) Graphics Card ASIN B01NALQ0PB |
| 7 | 4× AMD Radeon RX 7600 XT 16 GB | 12.9 GB | $2676.00 4 × $669.00 Amazon | GIGABYTE GAMING Radeon RX 7600 XT OC 16G AMD 16 GB GDDR6 ASIN B0CSSZCZZP |
| 8 | 4× AMD Radeon RX 9070 XT 16 GB | 12.9 GB | $2919.96 4 × $729.99 Amazon | GIGABYTE Radeon RX 9070 XT GAMING OC 16G Graphics Card ASIN B0DS2QG2KW |
| 9 | 4× NVIDIA Tesla T4 16 GB | 12.9 GB | $2936.00 4 × $734.00 Amazon | Lenovo 4X67A14926 graphics card NVIDIA Tesla T4 16 GB GDDR5X ASIN B07NSRK2DL |
| 10 | 3× AMD Radeon RX 7900 XTX 24 GB | 16.9 GB | $3179.97 3 × $1059.99 Amazon | GIGABYTE GAMING Radeon RX 7900 XTX OC 24G AMD 24 GB GDDR6 ASIN B0BRBXF3MW |
| 11 | 4× NVIDIA RTX 4080 16 GB | 12.9 GB | $3439.96 4 × $859.99 Amazon | GIGABYTE EAGLE GeForce RTX 4080 16GB OC NVIDIA GDDR6X ASIN B0BLT4ZMGL |
| 12 | 4× NVIDIA RTX 4070 Ti SUPER 16 GB | 12.9 GB | $3599.96 4 × $899.99 Amazon | MSI GeForce RTX 4070 Ti SUPER 16G VENTUS 2X WHITE OC NVIDIA 16 GB GDDR6X ASIN B0CSG9Z9ML |
| 13 | 4× NVIDIA RTX 5060 Ti 16 GB | 12.9 GB | $3627.88 4 × $906.97 Amazon | ASUS Dual -RTX5060TI-O16G NVIDIA GeForce RTX 5060 Ti 16 GB GDDR7 ASIN B0F4DVKSZ4 |
| 14 | 3× AMD Radeon RX 7900 XT 20 GB | 16.9 GB | $3659.61 3 × $1219.87 Amazon | ASUS TUF Gaming TUF-RX7900XT-O20G-GAMING AMD Radeon RX 7900 XT 20 GB GDDR6 ASIN B0BQCC78ZK |
| 15 | 4× AMD Radeon RX 6800 XT 16 GB | 12.9 GB | $3796.00 4 × $949.00 Amazon | MSI GAMING Radeon RX 6800 XT Z Trio 16G AMD 16 GB GDDR6 ASIN B09S596QVD |
| 16 | 4× AMD Radeon RX 7800 XT 16 GB | 12.9 GB | $3847.48 4 × $961.87 Amazon | GIGABYTE GAMING Radeon RX 7800 XT OC AMD 16 GB GDDR6 ASIN B0CGHQ32S2 |
| 17 | 3× NVIDIA RTX 3090 24 GB | 16.9 GB | $4041.66 3 × $1347.22 Amazon | GIGABYTE EAGLE GV-N3090EAGLE OC-24GD graphics card NVIDIA GeForce RTX 3090 24 GB GDDR6X ASIN B08HJPDJTY |
| 18 | 4× NVIDIA RTX 5070 Ti 16 GB | 12.9 GB | $4199.96 4 × $1049.99 Amazon | MSI GAMING GeForce RTX 5070 Ti 16G TRIO OC NVIDIA 16 GB GDDR7 ASIN B0F11KQQF4 |
| 19 | 4× AMD Radeon RX 7900 XTX 24 GB | 12.9 GB | $4239.96 4 × $1059.99 Amazon | GIGABYTE GAMING Radeon RX 7900 XTX OC 24G AMD 24 GB GDDR6 ASIN B0BRBXF3MW |
| 20 | 4× AMD Radeon RX 6900 XT 16 GB | 12.9 GB | $4589.12 4 × $1147.28 Amazon | MSI GAMING RX 6900 XT X TRIO 16G graphics card AMD Radeon RX 6900 XT 16 GB GDDR6 ASIN B08W2GPR62 |
Feasible — current price unavailable
- 4× NVIDIA Tesla M60 16 GB
- 4× AMD Radeon RX 9070 16 GB
- 4× AMD Radeon RX 9060 XT 16 GB
- 3× NVIDIA RTX 4500 24 GB
- 4× NVIDIA RTX 4500 24 GB
- 3× NVIDIA Tesla K80 24 GB
- 4× NVIDIA Tesla K80 24 GB
- 3× NVIDIA Tesla M40 24 GB
- 4× NVIDIA Tesla M40 24 GB
- 3× NVIDIA RTX 6000 24 GB
- 4× NVIDIA RTX 6000 24 GB
- 3× NVIDIA RTX A5500 24 GB
Query combinations are not indexed. Source: llama.cpp multi-GPU documentation.