daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-4bit-QAT-dequantized Text Generation • 0.4B • Updated about 18 hours ago
daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-4bit-QAT-dequantized Text Generation • 0.4B • Updated about 18 hours ago
GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling Paper • 2604.18556 • Published 25 days ago • 2
daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-2bit Text Generation • 41.1M • Updated 3 days ago • 31
daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-2bit Text Generation • 41.1M • Updated 3 days ago • 31
daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-3bit Text Generation • 54.8M • Updated 3 days ago • 38
daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-3bit Text Generation • 54.8M • Updated 3 days ago • 38
daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-4bit Text Generation • 68.5M • Updated 3 days ago • 30
daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-4bit Text Generation • 68.5M • Updated 3 days ago • 30
daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-5bit Text Generation • 82.2M • Updated 3 days ago • 34
daslab-testing/Apertus-0.6B-DPO-wnorm2both-MLX-5bit Text Generation • 82.2M • Updated 3 days ago • 34