A.Reverger
Reverger
AI & ML interests
None yet
Recent Activity
new activity 2 days ago
Kwaipilot/KAT-Coder-V2.5-Dev:Community benchmark results discussion new activity 8 days ago
Kwaipilot/KAT-Coder-V2.5-Dev:Please specify attention mechanism (how KV cache is treated)Organizations
None yet
Community benchmark results discussion
π 3
10
#12 opened 3 days ago
by
space-spike
Please specify attention mechanism (how KV cache is treated)
1
#2 opened 8 days ago
by
Reverger
Please specify max context size in model's card.
1
#8 opened 9 days ago
by
Reverger
this has been impressive in my usage so far, probably may become my daily driver
β€οΈ 4
1
#2 opened 2 months ago
by
AGmaton
Gemma 4 seems to work best with high temperature for coding
π 1
8
#21 opened 4 months ago
by
Reverger
Mar 5 - 'Final' Update: iMatrix + Benchmarks + New quant algo
β€οΈπ 7
30
#31 opened 5 months ago
by
danielhanchen
Feb 19: Qwen3-Coder-Next GGUFs update - much better outputs!
π€π 16
34
#5 opened 6 months ago
by
danielhanchen
Jan 21: GGUFs all UPDATED!!!
π€β€οΈ 5
40
#1 opened 6 months ago
by
danielhanchen
Please add Devstral-Small-2-24B-Instruct-2512 too
π 1
5
#4 opened 7 months ago
by
Reverger
Fixed system prompt + Updates
π 5
9
#5 opened 8 months ago
by
danielhanchen
Request GGUF: Kwaipilot/KAT-Dev (62.4% (!) on SWE-Bench Verified having just 32B)
π 1
1
#5 opened 10 months ago
by
Reverger
Is KAT-Dev supposed to be that much slower than QWEN?
4
#11 opened 10 months ago
by deleted
Request GGUF: Kwaipilot/KAT-Dev (62.4% (!) on SWE-Bench Verified having just 32B)
4
#15 opened 10 months ago
by
Reverger
why Kat-Dev worse then Qwen3-coder 30B?
12
#8 opened 10 months ago
by
akierum
Request GGUF: Kwaipilot/KAT-Dev (62.4% (!) on SWE-Bench Verified having just 32B)
4
#15 opened 10 months ago
by
Reverger
why Kat-Dev worse then Qwen3-coder 30B?
12
#8 opened 10 months ago
by
akierum