Qwen3.8-27B mixed-precision EXL3 (measured)
Collection
Four runtime-specific Qwen3.8 EXL3 builds. v5 10M shared-head, body-only KL vs BF16; advisory, not a quality ranking. Context tested separately. • 4 items • Updated • 1
🔆 Thinking....
Cosmic Nag
cosmicnag
AI & ML interests
None yet
Recent Activity
new activity about 4 hours ago
YTan2000/GLM-5.3-Flash-TQ3_4S:What exactly is using TQ compression liked a model 6 days ago
vcruz305/Qwen3.8-Flash-Next-EXL3-SAGE-4.15bpw liked a model 7 days ago
ukisai/Swift1.5-Qwen3.8-Flash-NextOrganizations
None yet
upvoted a collection about 1 month ago