REAP layerwise observations, calibration sets, and training data.
0xSero PRO
0xSero
AI & ML interests
Quantizing, benchmarking, training, and building.
Recent Activity
updated a model about 18 hours ago
0xSero/Kimi-K2.6-519B-NVFP4 published a model 2 days ago
0xSero/Kimi-K2.6-519B-NVFP4 updated a model 4 days ago
0xSero/DeepSeek-V4-Flash-Spark-GGUFOrganizations
Qwen — REAP
REAP-pruned & quantized Qwen3.5 / 3.6 / Coder variants.
Nemotron — REAP
REAP-pruned & quantized NVIDIA Nemotron-3-Super.
Gemma — REAP
REAP-pruned Gemma-4 MoE.
DeepSeek — REAP
REAP-pruned & quantized DeepSeek-V4-Flash / V3.2.
Trinity — REAP
REAP-pruned & quantized Trinity-Large-Thinking.
Other models
Hy3, Kimi-K2.5, INTELLECT-3, NousCoder.
MiniMax — REAP
REAP-pruned & quantized MiniMax-M2.1 / M2.7.
GLM — REAP
REAP-pruned & quantized GLM-4.x / 5 / 5.1 (+ Flash fine-tunes).
Proven REAPs
Benchmarked REAP checkpoints with >=500 all-time downloads. GLM/Qwen/MiniMax/DeepSeek/Kimi/gemma.
Datasets — observations & calibration
REAP layerwise observations, calibration sets, and training data.
Trinity — REAP
REAP-pruned & quantized Trinity-Large-Thinking.
Qwen — REAP
REAP-pruned & quantized Qwen3.5 / 3.6 / Coder variants.
Other models
Hy3, Kimi-K2.5, INTELLECT-3, NousCoder.
Nemotron — REAP
REAP-pruned & quantized NVIDIA Nemotron-3-Super.
MiniMax — REAP
REAP-pruned & quantized MiniMax-M2.1 / M2.7.
Gemma — REAP
REAP-pruned Gemma-4 MoE.
GLM — REAP
REAP-pruned & quantized GLM-4.x / 5 / 5.1 (+ Flash fine-tunes).
DeepSeek — REAP
REAP-pruned & quantized DeepSeek-V4-Flash / V3.2.
Proven REAPs
Benchmarked REAP checkpoints with >=500 all-time downloads. GLM/Qwen/MiniMax/DeepSeek/Kimi/gemma.