North Mini Code
CohereLabs/North-Mini-Code-1.0
Released2026-06-09Size30B · 3B activeArchitecturesliding-window attnLicenseapache-2.0AA rank#61 of 75
Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M bartowski/North-Mini-Code-1.0-GGUF | 17.46 GiB | 1 | 18.9 GiB | 20 GB GPU | HF files |
| Q5_K_M bartowski/North-Mini-Code-1.0-GGUF | 20.34 GiB | 1 | 21.8 GiB | 24 GB GPU | HF files |
| Q8_0 bartowski/North-Mini-Code-1.0-GGUF | 30.21 GiB | 1 | 31.7 GiB | 48 GB GPU | HF files |
| MLX 4-bit mlx-community/North-Mini-Code-1.0-4bit | 17.23 GiB | 4 | 18.7 GiB | 32 GB Mac | HF files |
| MLX 8-bit aaronmwelborn/North-Mini-Code-1.0-8bit | 30.98 GiB | 7 | 32.5 GiB | 48 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
9.9
Variant: North Mini Code · observed 2026-10-17 · source
Rank #61 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
- MLX chat_template.jinja Hugging Face · 2026-06-20
- llama.cpp? Hugging Face · 2026-06-09
Qualitative signals only; titles are not verified claims.