Qwopus3.6-27B-v2 MLX bf16

Full bf16 MLX reference of Jackrong/Qwopus3.6-27B-v2. No quantization — the original weights repacked into MLX format.

This is the master copy. Smaller quantized variants (4-bit, 6-bit, 8-bit, 3-bit, plus group-size sweeps) will be derived from the same source.

Use

pip install mlx-lm
mlx_lm.generate --model zaydiscold/Qwopus3.6-27B-v2-MLX-bf16 \
  --prompt "Explain quantum entanglement" --max-tokens 200 --temp 0.8

Author's temperature recommendation: 0.75–1.0.

Conversion

python -m mlx_lm convert --hf-path Jackrong/Qwopus3.6-27B-v2 --mlx-path ./output

Credits

Source: Jackrong/Qwopus3.6-27B-v2 by Jackrong
MLX conversion by zaydiscold

Mix brought by the NOTORIOUS MLX.

Downloads last month: 13

Safetensors

Model size

27B params

Tensor type

BF16

MLX

Hardware compatibility

Quantized

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for zaydiscold/Qwopus3.6-27B-v2-MLX-bf16

Base model

Jackrong/Qwopus3.6-27B-v2

Finetuned

(7)

this model

Collection including zaydiscold/Qwopus3.6-27B-v2-MLX-bf16

Qwopus3.6-27B-v2 MLX

Collection

MLX quant suite of Jackrong/Qwopus3.6-27B-v2 for Apple Silicon. Mix brought by the NOTORIOUS MLX. • 16 items • Updated May 23