Im a young AI/ML enthusiast collecting, training, and finetuning LLMs
-
gemma-4-claude-opus-4.6-thinking-s7-multimodal
Gemma 4 distilled from claude opus 4.6 thinking. Has only a 5% gap with claude opus 4.6 thinking while being over 40x smaller. Designed for server inference. Designed for local inference
tools thinking558 Pulls 1 Tag Updated 2 months ago
-
s5
A model based on the GLM-4.6v-flash:9b q5_k_m, and uncensored. For local use I recommend editing the model context in modelfile as it is set to 128k. #EDIT: New local optimised model same with context 4096 https://ollama.com/ShreyanGondaliya/s5-reduced
414 Pulls 1 Tag Updated 4 months ago
-
qwen-3.6-claude-opus-4.6-thinking-s8-pro-multimodal
A distil from opus 4.6 thinking. Only <7% gap in real world and benchmarking tests
392 Pulls 1 Tag Updated 1 month ago
-
s5-reduced
reduced s5 default context to 4096 tokens to allow better local inference thinks in chinese somtimes??
74 Pulls 1 Tag Updated 4 months ago