| RTX 4060 | Too big | | | | PDF: Qwen3.8 27B on RTX 4060 report |
|---|
| RTX 5060 | Too big | | | | PDF: Qwen3.8 27B on RTX 5060 report |
|---|
| RTX 3070 | Too big | | | | PDF: Qwen3.8 27B on RTX 3070 report |
|---|
| RTX 3080 10 GB | Too big | | | | PDF: Qwen3.8 27B on RTX 3080 10 GB report |
|---|
| RTX 3060 12 GB | Too big | | | | PDF: Qwen3.8 27B on RTX 3060 12 GB report |
|---|
| Arc B580 | Too big | | | | PDF: Qwen3.8 27B on Arc B580 report |
|---|
| RTX 4070 | Too big | | | | PDF: Qwen3.8 27B on RTX 4070 report |
|---|
| RTX 4070 Super | Too big | | | | PDF: Qwen3.8 27B on RTX 4070 Super report |
|---|
| RTX 5070 | Too big | | | | PDF: Qwen3.8 27B on RTX 5070 report |
|---|
| RTX 4060 Ti 16 GB | 3-bit | 20K | 15 tok/s | 2.5 s | PDF: Qwen3.8 27B on RTX 4060 Ti 16 GB report |
|---|
| RTX 5060 Ti 16 GB | 3-bit | 20K | 23 tok/s | 2.4 s | PDF: Qwen3.8 27B on RTX 5060 Ti 16 GB report |
|---|
| Arc A770 16 GB | 3-bit | 20K | 20 tok/s | 2.7 s | PDF: Qwen3.8 27B on Arc A770 16 GB report |
|---|
| Radeon RX 9070 XT | 3-bit | 20K | 27 tok/s | 1.6 s | PDF: Qwen3.8 27B on Radeon RX 9070 XT report |
|---|
| RTX 4070 Ti Super | 3-bit | 20K | 34 tok/s | 1.3 s | PDF: Qwen3.8 27B on RTX 4070 Ti Super report |
|---|
| RTX 4080 Super | 3-bit | 20K | 37 tok/s | 1.1 s | PDF: Qwen3.8 27B on RTX 4080 Super report |
|---|
| RTX 5070 Ti | 3-bit | 20K | 44 tok/s | 1.3 s | PDF: Qwen3.8 27B on RTX 5070 Ti report |
|---|
| RTX 5080 | 3-bit | 20K | 47 tok/s | 1.0 s | PDF: Qwen3.8 27B on RTX 5080 report |
|---|
| Radeon RX 7900 XTX | 5-bit | 48K | 28 tok/s | 2.5 s | PDF: Qwen3.8 27B on Radeon RX 7900 XTX report |
|---|
| RTX 3090 | 5-bit | 48K | 32 tok/s | 1.6 s | PDF: Qwen3.8 27B on RTX 3090 report |
|---|
| RTX 4090 | 5-bit | 48K | 35 tok/s | 0.7 s | PDF: Qwen3.8 27B on RTX 4090 report |
|---|
| Radeon AI PRO R9700 | 8-bit | 24K | 13 tok/s | 1.6 s | PDF: Qwen3.8 27B on Radeon AI PRO R9700 report |
|---|
| RTX 5090 | 8-bit | 24K | 41 tok/s | 0.5 s | PDF: Qwen3.8 27B on RTX 5090 report |
|---|
| 2× RTX 3090 | 8-bit | 256K | 22 tok/s | 0.9 s | PDF: Qwen3.8 27B on 2× RTX 3090 report |
|---|
| 2× RTX 4090 | 8-bit | 256K | 24 tok/s | 0.4 s | PDF: Qwen3.8 27B on 2× RTX 4090 report |
|---|
| RTX PRO 5000 Blackwell 48 GB | 8-bit | 256K | 31 tok/s | 0.8 s | PDF: Qwen3.8 27B on RTX PRO 5000 Blackwell 48 GB report |
|---|
| RTX 6000 Ada | 8-bit | 256K | 23 tok/s | 0.6 s | PDF: Qwen3.8 27B on RTX 6000 Ada report |
|---|
| L40S | 8-bit | 256K | 21 tok/s | 1.0 s | PDF: Qwen3.8 27B on L40S report |
|---|
| 2× RTX 5090 | 8-bit | 256K | 41 tok/s | 0.3 s | PDF: Qwen3.8 27B on 2× RTX 5090 report |
|---|
| 4× RTX 3090 | 8-bit | 256K | 22 tok/s | 0.5 s | PDF: Qwen3.8 27B on 4× RTX 3090 report |
|---|
| RTX PRO 6000 Blackwell | 8-bit | 256K | 41 tok/s | 0.4 s | PDF: Qwen3.8 27B on RTX PRO 6000 Blackwell report |
|---|
| Mac mini M4 16 GB | Too big | | | | PDF: Qwen3.8 27B on Mac mini M4 16 GB report |
|---|
| Mac mini M4 24 GB | 3-bit | 24K | 7.2 tok/s | 35 s | PDF: Qwen3.8 27B on Mac mini M4 24 GB report |
|---|
| Mac mini M4 32 GB | 4-bit | 56K | 5.9 tok/s | 35 s | PDF: Qwen3.8 27B on Mac mini M4 32 GB report |
|---|
| Mac mini M4 Pro 24 GB | 3-bit | 24K | 16 tok/s | 18 s | PDF: Qwen3.8 27B on Mac mini M4 Pro 24 GB report |
|---|
| Mac mini M4 Pro 48 GB | 8-bit | 88K | 7.6 tok/s | 18 s | PDF: Qwen3.8 27B on Mac mini M4 Pro 48 GB report |
|---|
| Mac mini M4 Pro 64 GB | 8-bit | 256K | 7.6 tok/s | 18 s | PDF: Qwen3.8 27B on Mac mini M4 Pro 64 GB report |
|---|
| M4 Max 128 GB | 8-bit | 256K | 13 tok/s | 8.9 s | PDF: Qwen3.8 27B on M4 Max 128 GB report |
|---|
| M5 Max 128 GB | 8-bit | 256K | 15 tok/s | 2.6 s | PDF: Qwen3.8 27B on M5 Max 128 GB report |
|---|
| M3 Ultra 96 GB | 8-bit | 256K | 17 tok/s | 5.0 s | PDF: Qwen3.8 27B on M3 Ultra 96 GB report |
|---|
| M3 Ultra 256 GB | 8-bit | 256K | 17 tok/s | 5.0 s | PDF: Qwen3.8 27B on M3 Ultra 256 GB report |
|---|
| M3 Ultra 512 GB | 8-bit | 256K | 17 tok/s | 5.0 s | PDF: Qwen3.8 27B on M3 Ultra 512 GB report |
|---|
| M5 Ultra 512 GB | 8-bit | 256K | 24 tok/s | 1.6 s | PDF: Qwen3.8 27B on M5 Ultra 512 GB report |
|---|
| Ryzen AI Max+ 395 64 GB | 8-bit | 256K | 5.3 tok/s | 5.1 s | PDF: Qwen3.8 27B on Ryzen AI Max+ 395 64 GB report |
|---|
| Ryzen AI Max+ 395 128 GB | 8-bit | 256K | 5.3 tok/s | 5.1 s | PDF: Qwen3.8 27B on Ryzen AI Max+ 395 128 GB report |
|---|
| RTX Spark | 8-bit | 256K | 7.3 tok/s | 1.8 s | PDF: Qwen3.8 27B on RTX Spark report |
|---|
| DGX Spark | 8-bit | 256K | 6.7 tok/s | 1.8 s | PDF: Qwen3.8 27B on DGX Spark report |
|---|
| Dual DGX Spark | 8-bit | 256K | 6.7 tok/s | 1.0 s | PDF: Qwen3.8 27B on Dual DGX Spark report |
|---|
| DGX Station GB300 | 8-bit | 256K | 131 tok/s | 0.1 s | PDF: Qwen3.8 27B on DGX Station GB300 report |
|---|
| 2× RTX PRO 6000 Blackwell | 8-bit | 256K | 41 tok/s | 0.2 s | PDF: Qwen3.8 27B on 2× RTX PRO 6000 Blackwell report |
|---|
| 4× RTX PRO 6000 Blackwell | 8-bit | 256K | 41 tok/s | 0.1 s | PDF: Qwen3.8 27B on 4× RTX PRO 6000 Blackwell report |
|---|
| 8× RTX PRO 6000 Blackwell | 8-bit | 256K | 41 tok/s | 0.1 s | PDF: Qwen3.8 27B on 8× RTX PRO 6000 Blackwell report |
|---|
| A100 80 GB SXM | 8-bit | 256K | 46 tok/s | 1.1 s | PDF: Qwen3.8 27B on A100 80 GB SXM report |
|---|
| H100 SXM | 8-bit | 256K | 71 tok/s | 0.4 s | PDF: Qwen3.8 27B on H100 SXM report |
|---|
| 4× H200 NVL | 8-bit | 256K | 97 tok/s | 0.1 s | PDF: Qwen3.8 27B on 4× H200 NVL report |
|---|
| 8× H200 NVL | 8-bit | 256K | 97 tok/s | 0.1 s | PDF: Qwen3.8 27B on 8× H200 NVL report |
|---|