KV-Cache Receipts — nine ≈128K users on one A100 (vLLM)
Collection
Three-arm, needle-gated serving receipts (fp16 / fp8-KV / compressed) — reproducible for ≈$12 with one command. • 5 items • Updated
AI & ML interests
KV cache compression, inference optimization, model compression
Recent Activity
View all activity
samuel-salfati
updated a
collection 6 days ago
samuel-salfati
published 2
models 6 days ago
samuel-salfati
updated 3
models 6 days ago
samuel-salfati
updated a
collection 7 days ago
samuel-salfati
published a
model 7 days ago
samuel-salfati
updated a
model 7 days ago
samuel-salfati
updated a
collection 10 days ago
samuel-salfati
published a
model 10 days ago
samuel-salfati
updated 8
models 10 days ago
fraQtl/Gemma-4-E2B-it-Hi-Fi-GGUF
Text Generation • 5B • Updated • 7.8k • 18
fraQtl/fraqtl-sm80-runtime
Updated • 11
fraQtl/mistral-7b-instruct-v0.3-kv-sidecars
Updated • 15
fraQtl/Qwen3.6-35B-A3B-Hi-Fi-MTP-runtime
Text Generation • 36B • Updated • 965 • 2
fraQtl/Llama-3.2-1B-Instruct-Hi-Fi-GGUF
Text Generation • 1B • Updated • 150
fraQtl/Llama-3.1-8B-Instruct-Hi-Fi-GGUF
Text Generation • 8B • Updated • 164
fraQtl/Gemma-4-26B-A4B-it-Hi-Fi-GGUF
Text Generation • 25B • Updated • 261 • 1
fraQtl/Qwen3.8-27B-Hi-Fi-GGUF
Text Generation • 27B • Updated • 298 • 2