MiniMax-M3
Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M bartowski/MiniMax-M3-GGUF | 243 GiB | 7 | unknown | over 192 GB GPU | HF files |
| Q5_K_M bartowski/MiniMax-M3-GGUF | 284 GiB | 8 | unknown | over 192 GB GPU | HF files |
| Q8_0 unsloth/MiniMax-M3-GGUF | 422 GiB | 11 | unknown | over 192 GB GPU | HF files |
| BF16 unsloth/MiniMax-M3-GGUF | 794 GiB | 18 | unknown | over 192 GB GPU | HF files |
| MLX 4-bit mixed precision mlx-community/MiniMax-M3-4bit | 225 GiB | 58 | unknown | 512 GB Mac (cache unknown) | HF files |
| MLX 8-bit mixed precision pipenetwork/MiniMax-M3-MLX-8bit | 421 GiB | 87 | unknown | over 512 GB Mac | HF files |
No verified file: FP16. 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.
Quality
Artificial Analysis Intelligence Index
29.2
Variant: MiniMax-M3 · observed 2026-10-17 · source
Rank #17 of 75 researched 2026 models.
Arena text rating
1,433
95% CI 1430–1437 · 63,973 votes · published 2026-10-15 · dataset
Exact listed model name; hosted reasoning/quantization may differ from local checkpoint.
Scores describe the evaluated hosted variant, not each quantized file. AA and Arena are never averaged.
Reported speeds 6
| Hardware | Quant | Backend | Prompt t/s | Gen t/s | Date | Source |
|---|---|---|---|---|---|---|
| Apple M3 Ultra (80-core GPU), 512 GB unified memory | 4bit (MiniMax-M3-4bit) | MLX (oMLX server) | 343 | 26.9 | 2026-08-25 | omlx.ai |
| Apple M3 Ultra (80-core GPU), 512 GB unified memory | 4bit (MiniMax-M3-4bit) | MLX (oMLX server) | 339 | 19.1 | 2026-08-25 | omlx.ai |
| Apple M3 Ultra (80-core GPU), 512 GB unified memory | 4bit (MiniMax-M3-4bit) | MLX (oMLX server) | 313 | 18.3 | 2026-08-25 | omlx.ai |
| Apple M3 Ultra (80-core GPU), 512 GB unified memory | 4bit (MiniMax-M3-4bit) | MLX (oMLX server) | 276 | 5.3 | 2026-08-25 | omlx.ai |
| Apple M3 Ultra (80-core GPU), 512 GB unified memory | 4bit (MiniMax-M3-4bit) | MLX (oMLX server) | 329 | 19 | 2026-08-25 | omlx.ai |
| Apple M3 Ultra (80-core GPU), 256 GB unified memory | 3bit (MiniMax-M3-Alis-MLX-Dynamic) | MLX (oMLX server) | 219 | 19 | 2026-08-13 | omlx.ai |
Reported by the linked sources, not reproduced by LocalFit. All rows in the speed database