Policies trained in simulation for a Pollen Robotics Microduck still on order, one level per skill, plus the training log. Sim only so far.
🏗️ Building on HF
witcheer
witcheer
P(doom) 10%
AI & ML interests
Local AI Maxxing.
Recent Activity
updated a dataset 1 day ago
witcheer/rtx-5090-benchmarks updated a collection 6 days ago
Microduck skill tree (sim only) updated a dataset 6 days ago
witcheer/microduck-skill-treeOrganizations
Policies trained in simulation for a Pollen Robotics Microduck still on order, one level per skill, plus the training log. Sim only so far.
Speed and quality benchmarks for quantized LLMs on NVIDIA RTX 5090 32GB. Data, tooling, and results.
11 models on RTX 4060 Ti 8GB. speed king: Llama 1B (228 tok/s). quality king: Gemma 4 E4B (6/6). agentic king: gpt-oss-20b (10/10).