Mudit Gupta @Mudit__Gupta · 26 Aug 2026
This is by far the best local llm model you can run on a 128GB Mac. It has only 6b active parameters making it super fast even on unified memory! And it's intelligent also :) Can't wait for them to launch a 27b/35b model on this architecture so we can run it on 36GB Macs
4 934Views
35Likes
2Reposts
2Replies
0Quotes
16Bookmarks
Is that a lot?
1.35×vs this author's median3 665 views is typical
56Percentile for this authorof 9 recent posts
2.53×vs 10K–100K median1 948 views is typical
7.32%Reachviews ÷ followers
0.79%Engagement rateof viewers reacted