Running
Qwen3.6 llama.cpp TurboQuant Smoke Results
🚀
View performance benchmarks for Qwen3.6 LLM runtimes
Jake
sjakek
AI & ML interests
MoE tuning + optimizing, offloading
Recent Activity
new activity 7 days ago
sjakek/Nex-N2-mini-GGUF:Unsloth-style UD quants) updated a model 7 days ago
sjakek/Nex-N2-mini-GGUF published a model 7 days ago
sjakek/Nex-N2-mini-GGUFOrganizations
None yet
datasets 10
sjakek/qwopus-dflash-swe20-runtime-results
Viewer • Updated • 14.7k • 227
sjakek/lfm25-hermes-tool-trace-analysis
Preview • Updated • 222
sjakek/qwen36-runtime-analysis-3090ti
Updated • 340
sjakek/qwen36-kquant-offload-mtp-swebench-lite100-results
Viewer • Updated • 1 • 135
sjakek/qwen36-coding-layer-bottleneck-profiles
Updated • 298
sjakek/qwen36-q4km-swebench-lite-saliency-overfit-layercompare
Preview • Updated • 103
sjakek/qwen36-q4km-saliency-vs-attention-residency-100
Preview • Updated • 108
sjakek/qwen36-moe-coding-expert-saliency
Updated • 98
sjakek/qwen36-mtp-turbo-kv-analysis
Viewer • Updated • 7 • 427 • 1
sjakek/qwen36-27b-mtp-long-context-decay
Updated • 206