Turkish LLM Family
Şuayp Talha Kocabay
suayptalha
AI & ML interests
NLP, LLMs, Transformers, Merging, RNNs, CNNs, ANNs, Computer Vision and ML algorithms
Recent Activity
upvoted a paper 9 days ago
Efficient Pre-Training with Token Superposition authored a paper about 1 month ago
Selectivity and Shape in the Design of Forward-Forward Goodness Functions liked a model about 1 month ago
moonshotai/Kimi-K2.6Organizations
Maestro LLMs based on DeepSeek's distilled models
A Faster and Higher-performing FastLlama Series
Monoalphabetic Substitution Cipher Solver Models
Qwen3 MoE model
-
suayptalha/Arcana-Qwen3-2.4B-A0.6B
Text Generation • Updated • 17 • 33 -
suayptalha/Qwen3-0.6B-Medical-Expert
Text Generation • 0.6B • Updated • 32 • • 5 -
suayptalha/Qwen3-0.6B-Code-Expert
Text Generation • 0.6B • Updated • 881 • • 9 -
suayptalha/Qwen3-0.6B-Math-Expert
Text Generation • 0.6B • Updated • 73 • • 1
Hugging Face integration of minGRU RNN models
Models I merged using mergekit library
-
suayptalha/Lamarckvergence-14B
Text Generation • 15B • Updated • 47 • 27 -
suayptalha/Luminis-phi-4
Text Generation • 15B • Updated • 1.11k • 13 -
suayptalha/Falcon3-Jessi-v0.4-7B-Slerp
Text Generation • 7B • Updated • 15 • 9 -
suayptalha/HomerCreativeAnvita-Mix-Qw7B
Text Generation • 8B • Updated • 17 • • 12
Komodo LLM Series
Turkish LLM Family
Qwen3 MoE model
-
suayptalha/Arcana-Qwen3-2.4B-A0.6B
Text Generation • Updated • 17 • 33 -
suayptalha/Qwen3-0.6B-Medical-Expert
Text Generation • 0.6B • Updated • 32 • • 5 -
suayptalha/Qwen3-0.6B-Code-Expert
Text Generation • 0.6B • Updated • 881 • • 9 -
suayptalha/Qwen3-0.6B-Math-Expert
Text Generation • 0.6B • Updated • 73 • • 1
Maestro LLMs based on DeepSeek's distilled models
Hugging Face integration of minGRU RNN models
A Faster and Higher-performing FastLlama Series
Models I merged using mergekit library
-
suayptalha/Lamarckvergence-14B
Text Generation • 15B • Updated • 47 • 27 -
suayptalha/Luminis-phi-4
Text Generation • 15B • Updated • 1.11k • 13 -
suayptalha/Falcon3-Jessi-v0.4-7B-Slerp
Text Generation • 7B • Updated • 15 • 9 -
suayptalha/HomerCreativeAnvita-Mix-Qw7B
Text Generation • 8B • Updated • 17 • • 12
Monoalphabetic Substitution Cipher Solver Models
Komodo LLM Series