Comment by verdverm 6 hours ago Similar, but I'm using 35B A3B variation with experimental MTP supportOpenCode is pretty good too 2 comments verdverm Reply danielbln 6 hours ago A3B is especially nice, MoE really shines on memory bandwidth contained platforms like the DGX Spark. verdverm 3 hours ago looks like MTP support has now been merged and also updated unsloth quants to go with it (not just the extras, all of 'em!)
danielbln 6 hours ago A3B is especially nice, MoE really shines on memory bandwidth contained platforms like the DGX Spark. verdverm 3 hours ago looks like MTP support has now been merged and also updated unsloth quants to go with it (not just the extras, all of 'em!)
verdverm 3 hours ago looks like MTP support has now been merged and also updated unsloth quants to go with it (not just the extras, all of 'em!)
A3B is especially nice, MoE really shines on memory bandwidth contained platforms like the DGX Spark.
looks like MTP support has now been merged and also updated unsloth quants to go with it (not just the extras, all of 'em!)