abenzerps/MiniCPM5-2B-MLX-8bit

Verified creator abenzerps verified
🤗 Hugging Face sourcetext-generationapache-2.02.5B params5.0 GBsafetensors✓ 1 checksumupdated today
Needs seeder →

MiniCPM5-2B MLX - 8-bit

MLX 8-bit conversion of OpenBMB/MiniCPM5-2B, a 2B dense Llama-based model for local deployment, coding, reasoning, long-context work, and tool use. The source checkpoint supports a native context length of 131,072 tokens (128K).

Benchmarks

Benchmark results reported by OpenBMB for the original MiniCPM5-2B checkpoint.

Release

Format Quantization Size
MLX safetensors Affine 8-bit, group size 64 2.5 GB

The model is text-only. No vision projector or MTP files are included.

Usage

pip install -U mlx-lm
mlx_lm.generate \
  --model . \
  --prompt "Explain why reproducible builds matter." \
  --max-tokens 512

Source