A kitchen-sink collection of experimental auxiliary-head architectures built on DeepSeek Flash, exploring anything that might extend MOE architectures
Nicholai Mitchko
nmitchko
AI & ML interests
Fine-tuning, Scaling, Enablement, Activation Engineering
Recent Activity
upvoted a paper about 24 hours ago
Think Silently, Think Fast: Dynamic Latent Compression of LLM Reasoning
Chains updated a model 10 days ago
nmitchko/deepseek-v4-Flash-CoLaR published a model 14 days ago
nmitchko/deepseek-v4-Flash-CoLaROrganizations
None yet
models 12
nmitchko/deepseek-v4-Flash-CoLaR
Text Generation • Updated • 469
nmitchko/medguanaco-65b-GPTQ
Text Generation • Updated • 6 • 5
nmitchko/dr-niko-lora-70B
Updated • 1
nmitchko/medfalcon-v2-40b-lora
Text Generation • Updated • 12 • 3
nmitchko/dr-niko-70B
Text Generation • 69B • Updated • 7 • 1
nmitchko/ML1-previews
Text Generation • Updated • 4 • 4
nmitchko/i2b2-querybuilder-34b-merged
Text Generation • 34B • Updated • 2
nmitchko/i2b2-querybuilder-codellama-34b
Text Generation • Updated • 5
nmitchko/medfalconv2-1a-40b-lora
Text Generation • Updated • 2
nmitchko/medfalcon-40b-lora
Text Generation • Updated • 3 • 3