Michael Druggan @Michael_Druggan · 31 Aug 2026
Today we're releasing abliterated-model-large-v2. Based on GLM-5.3, which is #3 on Terminal-Bench 4.0 (behind only Opus 5 and Fable), with 2× the cyber exploitation of 5.2. We abliterated and hosted it so it does the offensive cyber, red teaming, and agent testing work other https://t.co/rVRhcEEgfI
2 647 774Views
6 901Likes
435Reposts
288Replies
453Quotes
0Bookmarks
Is that a lot?
268×vs this author's median9 890 views is typical
92Percentile for this authorof 13 recent posts
1 359×vs 10K–100K median1 948 views is typical
167.3× audienceReachviews ÷ followers
0.30%Engagement rateof viewers reacted