AI & ML interests
Hardware-aware AI Model Optimization
Recent Activity
View all activity
Compressed NVIDIA Models (Nemotron and Cosmos)
Solar Open models officially optimized by Nota AI for Korea’s government-led Sovereign AI initiative as part of the Upstage consortium.
-
nota-ai/Solar-Open2-250B-Nota-INT4
Text Generation • 41B • Updated • 1.86k • 40 -
nota-ai/Solar-Open2-250B-Nota-NVFP4
Text Generation • 145B • Updated • 11.1k • 177 -
nota-ai/Solar-Open2-250B-Nota-INT4-GlobalPruned
Text Generation • 35B • Updated • 127 • 43 -
nota-ai/Solar-Open2-250B-Nota-NVFP4-GlobalPruned
Text Generation • 117B • Updated • 118 • 34
ERGO: LVLM trained with RL on efficiency objectives; https://github.com/nota-github/ERGO
Shortened LLMs from Depth Pruning; https://github.com/Nota-NetsPresso/shortened-llm
-
nota-ai/cpt_st-vicuna-v1.3-1.5b-ppl
Text Generation • 1B • Updated • 14 • 4 -
nota-ai/cpt_st-vicuna-v1.3-2.7b-ppl
Text Generation • 3B • Updated • 13 • 5 -
nota-ai/cpt_st-vicuna-v1.3-3.7b-ppl
Text Generation • 4B • Updated • 17 • 4 -
nota-ai/cpt_st-vicuna-v1.3-5.5b-ppl
Text Generation • 6B • Updated • 12 • 4
-
nota-ai/Qwen3.8-Flash-Next-Nota-NVFP4
Image-Text-to-Text • 180B • Updated • 451 • 21 -
nota-ai/Qwen3.8-2.4T-A95B-Nota-NVFP4-Global-Pruned-40
Text Generation • 1.5T • Updated • 569 • 23 -
nota-ai/Qwen3-30B-A3B-NotaMoEQuant-Int4
Text Generation • 0.6B • Updated • 30 • 8 -
nota-ai/Qwen3.5-122B-A10B-NotaCompression-INT4
Text Generation • 22B • Updated • 279 • 1
Optimized Kimi models
Mixture-of-Experts Large Language Models with Advanced Quantization
-
nota-ai/Solar-Open-100B-NotaMoEQuant-NVFP4
Text Generation • 59B • Updated • 233 • 23 -
nota-ai/Solar-Open-100B-Nota-FP8
Text Generation • 103B • Updated • 107 • 46 -
nota-ai/Solar-Open-100B-NotaMoEQuant-Int4
Text Generation • 2B • Updated • 284 • 62 -
nota-ai/Qwen3-30B-A3B-NotaMoEQuant-Int4
Text Generation • 0.6B • Updated • 30 • 8
Block-removed Knowledge-distilled SD models; https://github.com/Nota-NetsPresso/BK-SDM
-
nota-ai/Qwen3.8-Flash-Next-Nota-NVFP4
Image-Text-to-Text • 180B • Updated • 451 • 21 -
nota-ai/Qwen3.8-2.4T-A95B-Nota-NVFP4-Global-Pruned-40
Text Generation • 1.5T • Updated • 569 • 23 -
nota-ai/Qwen3-30B-A3B-NotaMoEQuant-Int4
Text Generation • 0.6B • Updated • 30 • 8 -
nota-ai/Qwen3.5-122B-A10B-NotaCompression-INT4
Text Generation • 22B • Updated • 279 • 1
Compressed NVIDIA Models (Nemotron and Cosmos)
Optimized Kimi models
Solar Open models officially optimized by Nota AI for Korea’s government-led Sovereign AI initiative as part of the Upstage consortium.
-
nota-ai/Solar-Open2-250B-Nota-INT4
Text Generation • 41B • Updated • 1.86k • 40 -
nota-ai/Solar-Open2-250B-Nota-NVFP4
Text Generation • 145B • Updated • 11.1k • 177 -
nota-ai/Solar-Open2-250B-Nota-INT4-GlobalPruned
Text Generation • 35B • Updated • 127 • 43 -
nota-ai/Solar-Open2-250B-Nota-NVFP4-GlobalPruned
Text Generation • 117B • Updated • 118 • 34
Mixture-of-Experts Large Language Models with Advanced Quantization
-
nota-ai/Solar-Open-100B-NotaMoEQuant-NVFP4
Text Generation • 59B • Updated • 233 • 23 -
nota-ai/Solar-Open-100B-Nota-FP8
Text Generation • 103B • Updated • 107 • 46 -
nota-ai/Solar-Open-100B-NotaMoEQuant-Int4
Text Generation • 2B • Updated • 284 • 62 -
nota-ai/Qwen3-30B-A3B-NotaMoEQuant-Int4
Text Generation • 0.6B • Updated • 30 • 8
ERGO: LVLM trained with RL on efficiency objectives; https://github.com/nota-github/ERGO
Block-removed Knowledge-distilled SD models; https://github.com/Nota-NetsPresso/BK-SDM
Shortened LLMs from Depth Pruning; https://github.com/Nota-NetsPresso/shortened-llm
-
nota-ai/cpt_st-vicuna-v1.3-1.5b-ppl
Text Generation • 1B • Updated • 14 • 4 -
nota-ai/cpt_st-vicuna-v1.3-2.7b-ppl
Text Generation • 3B • Updated • 13 • 5 -
nota-ai/cpt_st-vicuna-v1.3-3.7b-ppl
Text Generation • 4B • Updated • 17 • 4 -
nota-ai/cpt_st-vicuna-v1.3-5.5b-ppl
Text Generation • 6B • Updated • 12 • 4