Llama 4 Scout 17B-16E Instruct Legacy
Files and memory at 8K context
No verified quantized file for the requested formats.
Sizes are exact file bytes (text weights; projectors listed separately). “+” = recurrent state or other terms unknown. Smallest fit adds 0.5 GiB spare; Macs use 75% of memory. Legacy model: sizes on device pages are estimated from parameter count.
Quality
Artificial Analysis Intelligence Index
Unknown
Arena text rating
Unknown
Scores describe the evaluated hosted variant, not each quantized file. AA and Arena are never averaged.
Reported speeds 10
| Hardware | Quant | Backend | Prompt t/s | Gen t/s | Date | Source |
|---|---|---|---|---|---|---|
| Apple M3 Ultra (60-core GPU), 256 GB unified memory | MLX 6-bit | MLX | 455 | 32 | 2026-05-20 | omlx.ai |
| Framework Desktop (AMD Ryzen AI Max+ 395, 128GB LPDDR5x-8000, pre-production) | UD-Q4_K_XL | llama.cpp ROCm | 277 | 17.6 | 2025-08-08 | lhl/strix-halo-testing |
| Framework Desktop (AMD Ryzen AI Max+ 395, 128GB LPDDR5x-8000, pre-production) | UD-Q4_K_XL | llama.cpp Vulkan | 158 | 19.4 | 2025-08-08 | lhl/strix-halo-testing |
| Framework Desktop (AMD Ryzen AI Max+ 395, 128GB LPDDR5x-8000, pre-production) | UD-Q4_K_XL | llama.cpp ROCm | 264 | 17.2 | 2025-07-22 | forum.level1techs.com |
| Framework Desktop (AMD Ryzen AI Max+ 395, 128GB LPDDR5x-8000, pre-production) | UD-Q4_K_XL | llama.cpp Vulkan | 146 | 19.3 | 2025-07-22 | forum.level1techs.com |
| M3 Ultra 512GB | MLX 3-bit | MLX | — | 52.9 | 2025-04-06 | mlx-lm #74 |
| M3 Ultra 512GB | MLX 4-bit | MLX | — | 46.9 | 2025-04-06 | mlx-lm #74 |
| M3 Ultra 512GB | MLX 6-bit | MLX | — | 36.3 | 2025-04-06 | mlx-lm #74 |
| M3 Ultra 512GB | MLX 8-bit | MLX | — | 30.4 | 2025-04-06 | mlx-lm #74 |
| M3 Ultra 512GB | MLX fp16 | MLX | — | 11.7 | 2025-04-06 | mlx-lm #74 |
Reported by the linked sources, not reproduced by LocalFit. All rows in the speed database