Hacker News new | ask | show | jobs
by verdverm 35 days ago
Similar, but I'm using 35B A3B variation with experimental MTP support

OpenCode is pretty good too
1 comments
danielbln 35 days ago
A3B is especially nice, MoE really shines on memory bandwidth contained platforms like the DGX Spark.
link
verdverm 34 days 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!)
link