Llama · Apr 2025 · 109B total, 17B active
Llama 4 Scout 109B-A17B: what runs it and what it costs today
At IQ2_M with 8K of context it needs about 40 GB of memory. The cheapest machine that runs it today is the Framework Desktop, Ryzen AI Max+ 395, 128 GB at $3,464.65.
By quantization
Memory needed includes the KV cache for 8K of context and runtime overhead. The cheapest machine is the lowest live price, item plus shipping, that fits with room for that context.
| Quant | Needs | Cheapest machine that fits today | Price |
|---|---|---|---|
| FP16 | 221 GB | Apple Mac Studio M3 Ultra, 256 GB | $8,499 |
| Q8_0 | 119 GB | Framework Desktop, Ryzen AI Max+ 395, 128 GB | $3,464.65 |
| Q6_K | 92 GB | Framework Desktop, Ryzen AI Max+ 395, 128 GB | $3,464.65 |
| Q5_K_M | 81 GB | Framework Desktop, Ryzen AI Max+ 395, 128 GB | $3,464.65 |
| Q4_K_M | 70 GB | Framework Desktop, Ryzen AI Max+ 395, 128 GB | $3,464.65 |
| Q3_K_M | 57 GB | Framework Desktop, Ryzen AI Max+ 395, 128 GB | $3,464.65 |
| IQ2_M | 40 GB | Framework Desktop, Ryzen AI Max+ 395, 128 GB | $3,464.65 |
Every machine at IQ2_M 16 fit
| Machine | Memory | Fit at 8K | Est. speed | Price | Per GB |
|---|---|---|---|---|---|
| Framework Desktop, Ryzen AI Max+ 395, 128 GB desktop box | 128 GB | fits 128K+ ctx | ~22 tok/s | $3,464.65 eBay | $27/GB |
| GMKtec EVO-X2, Ryzen AI Max+ 395, 128 GB desktop box | 128 GB | fits 128K+ ctx | ~22 tok/s | $3,649.99 eBay | $28/GB |
| NVIDIA RTX A6000 card | 48 GB | fits 46K ctx | ~86 tok/s | $4,500 eBay | $93/GB |
| Apple Mac Studio M4 Max, 128 GB desktop box | 128 GB | fits 128K+ ctx | ~57 tok/s | $4,899 eBay | $38/GB |
| ASUS Ascent GX10 desktop box | 128 GB | fits 128K+ ctx | ~31 tok/s | $6,099.99 eBay | $47/GB |
| Acer Veriton GN100 desktop box | 128 GB | fits 128K+ ctx | ~31 tok/s | $6,367.19 eBay | $49/GB |
| HP ZGX Nano AI Station desktop box | 128 GB | fits 128K+ ctx | ~31 tok/s | $6,440.40 eBay | $50/GB |
| Gigabyte AI TOP ATOM desktop box | 128 GB | fits 128K+ ctx | ~31 tok/s | $6,999 eBay | $54/GB |
| NVIDIA RTX PRO 5000 Blackwell card | 48 GB | fits 46K ctx | ~150 tok/s | $7,099.99 eBay | $147/GB |
| NVIDIA RTX 6000 Ada card | 48 GB | fits 46K ctx | ~110 tok/s | $7,698.43 eBay | $160/GB |
| NVIDIA DGX Spark desktop box | 128 GB | fits 128K+ ctx | ~31 tok/s | $7,799.98 eBay | $60/GB |
| Apple Mac Studio M3 Ultra, 256 GB desktop box | 256 GB | fits 128K+ ctx | ~85 tok/s | $8,499 eBay | $33/GB |
| NVIDIA A100 80GB PCIe card | 80 GB | fits 128K+ ctx | ~220 tok/s | $15,299 eBay | $191/GB |
| NVIDIA RTX PRO 6000 Blackwell card | 96 GB | fits 128K+ ctx | ~200 tok/s | $16,499 eBay | $171/GB |
| Apple Mac Studio M3 Ultra, 512 GB desktop box | 512 GB | fits 128K+ ctx | ~85 tok/s | $17,499 eBay | $34/GB |
| NVIDIA H100 80GB PCIe card | 80 GB | fits 128K+ ctx | ~220 tok/s | $26,995.21 eBay | $337/GB |
| NVIDIA Tesla P100 16GB card | 16 GB | does not fit | $127.63 eBay | $7/GB | |
| AMD Radeon Instinct MI50 16GB card | 16 GB | does not fit | $228.78 eBay | $14/GB | |
| NVIDIA Tesla P40 card | 24 GB | does not fit | $273 eBay | $11/GB | |
| NVIDIA GeForce RTX 3060 12GB card | 12 GB | does not fit | $320 eBay | $26/GB | |
| AMD Radeon Instinct MI50 32GB card | 32 GB | does not fit | $600 eBay | $18/GB | |
| NVIDIA GeForce RTX 5060 Ti 16GB card | 16 GB | does not fit | $617.99 eBay | $38/GB | |
| NVIDIA GeForce RTX 4060 Ti 16GB card | 16 GB | does not fit | $656.27 eBay | $41/GB | |
| AMD Radeon RX 7900 XTX card | 24 GB | does not fit | $900 eBay | $37/GB | |
| NVIDIA RTX A4000 card | 16 GB | does not fit | $950 eBay | $59/GB | |
| NVIDIA GeForce RTX 3090 card | 24 GB | does not fit | $1,420 eBay | $59/GB | |
| NVIDIA RTX 4000 SFF Ada card | 20 GB | does not fit | $1,510.30 eBay | $75/GB | |
| AMD Radeon AI PRO R9700 card | 32 GB | does not fit | $1,820 eBay | $56/GB | |
| NVIDIA GeForce RTX 4090 card | 24 GB | does not fit | $2,083.48 eBay | $86/GB | |
| NVIDIA RTX A5000 card | 24 GB | does not fit | $2,325.65 eBay | $96/GB | |
| NVIDIA RTX PRO 4000 Blackwell card | 24 GB | does not fit | $2,827.16 eBay | $117/GB | |
| NVIDIA RTX PRO 4000 Blackwell SFF card | 24 GB | does not fit | $2,899 eBay | $120/GB | |
| NVIDIA RTX PRO 4500 Blackwell card | 32 GB | does not fit | $4,450 eBay | $139/GB | |
| NVIDIA GeForce RTX 5090 card | 32 GB | does not fit | $4,920.99 eBay | $153/GB |
Estimates assume the whole model in memory with llama.cpp-style inference; see how we estimate. Prices are the cheapest live eBay listing that passed our filters, item plus shipping.