LFM2.5-24B-A2B
Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M LiquidAI/LFM2.5-24B-A2B-GGUF | 13.53 GiB | 1 | 14.7 GiB+ | 16 GB GPU | HF files |
| Q5_K_M LiquidAI/LFM2.5-24B-A2B-GGUF | 15.86 GiB | 1 | 17.0 GiB+ | 20 GB GPU | HF files |
| Q8_0 LiquidAI/LFM2.5-24B-A2B-GGUF | 23.74 GiB | 1 | 24.9 GiB+ | 32 GB GPU | HF files |
| FP16 LiquidAI/LFM2.5-24B-A2B-GGUF | 44.67 GiB | 1 | 45.8 GiB+ | 48 GB GPU | HF files |
| MLX 4-bit mixed precision LiquidAI/LFM2.5-24B-A2B-MLX-4bit | 12.64 GiB | 3 | 13.8 GiB+ | 24 GB Mac | HF files |
| MLX 8-bit LiquidAI/LFM2.5-24B-A2B-MLX-8bit | 23.73 GiB | 5 | 24.9 GiB+ | 36 GB Mac | HF files |
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
Unknown
Arena text rating
Unknown
Scores describe the evaluated hosted variant, not each quantized file. AA and Arena are never averaged.
Reported speeds 5
| Hardware | Quant | Backend | Prompt t/s | Gen t/s | Date | Source |
|---|---|---|---|---|---|---|
| M4 Max (40c GPU) 128 GB | MLX 4bit | MLX | 2,557 | 151 | 2026-10-16 | omlx.ai |
| M5 Pro (20c GPU) 48 GB | MLX 4bit | MLX | 2,686 | 122 | 2026-10-16 | omlx.ai |
| M4 Pro (16c GPU) 48 GB | MLX 4bit | MLX | 1,028 | 107 | 2026-10-14 | omlx.ai |
| M4 Pro (20c GPU) 48 GB | MLX 4bit | MLX | 1,260 | 104 | 2026-10-14 | omlx.ai |
| M1 Max (24c GPU) 64 GB | MLX 4bit | MLX | 755 | 71.3 | 2026-10-13 | omlx.ai |
Reported by the linked sources, not reproduced by LocalFit. All rows in the speed database
Community threads
- LFM2.5-24B-A2B, vram requirements Hugging Face · 2026-10-14
- Blog post error Hugging Face · 2026-10-13
Qualitative signals only; titles are not verified claims.