llmtech/decider-4b-fp8
Text Classification • 4B • Updated • 60 • 2
artem
fogf
AI & ML interests
None yet
Recent Activity
new activity 17 days ago
Mapika/decider-0.8b:FP8 checkpoint for vLLM, measured against bf16 new activity 17 days ago
Mapika/decider-2b:FP8 checkpoint for vLLM, measured against bf16 new activity 17 days ago
Mapika/decider-4b:FP8 and NVFP4 checkpoints for vLLM, measured against bf16Organizations
FP8 checkpoint for vLLM, measured against bf16
#2 opened 17 days ago
by fogf
FP8 checkpoint for vLLM, measured against bf16
#3 opened 17 days ago
by fogf
FP8 and NVFP4 checkpoints for vLLM, measured against bf16
#2 opened 17 days ago
by fogf
liked a
model 17 days ago
updated a
Space 17 days ago
published a
Space 17 days ago
updated 2
models 17 days ago
liked 3
models 17 days ago
updated 2
models 17 days ago
published 4
models 17 days ago
Subject: NVFP4 quant of decider-0.8b/2b/4b
#1 opened 17 days ago
by fogf
Serving notes: prefix cache on the hybrid attention (1584-token blocks, lazy materialization) + thinking field gotchas
👍 3
1
#169 opened about 2 months ago
by fogf
liked a
model about 2 months ago
