Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M openbmb/MiniCPM5-2B-GGUF | 1.45 GiB | 1 | 2.8 GiB | 8 GB GPU | HF files |
| Q5_K_M bartowski/MiniCPM5-2B-GGUF | 1.79 GiB | 1 | 3.1 GiB | 8 GB GPU | HF files |
| Q8_0 openbmb/MiniCPM5-2B-GGUF | 2.50 GiB | 1 | 3.8 GiB | 8 GB GPU | HF files |
| FP16 openbmb/MiniCPM5-2B-GGUF | 4.69 GiB | 1 | 6.0 GiB | 8 GB GPU | HF files |
| MLX 4-bit openbmb/MiniCPM5-2B-MLX | 1.32 GiB | 1 | 2.6 GiB | 8 GB Mac | HF files |
| MLX 8-bit mlx-community/MiniCPM5-2B-8bit | 2.49 GiB | 1 | 3.8 GiB | 8 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
12.5
Variant: MiniCPM5-2B · observed 2026-10-17 · source
Rank #51 of 75 researched 2026 models.
Arena text rating
Unknown
Scores describe the evaluated hosted variant, not each quantized file. AA and Arena are never averaged.
Reported speeds 0
No reported measurement yet. Submit one.
Community threads
- MiniCPM5-2B-Q8_0_XL-GGUF Hugging Face · 2026-09-26
- it will be good if model supports turkish Hugging Face · 2026-09-08
Qualitative signals only; titles are not verified claims.