the problem is the dsv4 checkpoint so quantized isn't very good
Which ds4 checkpoint for which model exactly did you test? Don't they have multiple different versions and quantization levels?
not my experience
the ds4 quants were very good beating the unsloth quants https://github.com/michaelasper/benchmarks/blob/main/deepsee...
Which ds4 checkpoint for which model exactly did you test? Don't they have multiple different versions and quantization levels?