Open research artifacts, four-axis carded.
Every artifact here ships with the same disclosure shape — quality, throughput, vertical skill, thermal envelope — and was measured on a single DGX Spark. Browse by kind.
Quantized base models, Spark-measured across perplexity, throughput, vertical eval, and thermal envelope.
Browse quantizationsLoRA adapters with training receipts and inference deltas.
Prefix, prompt, and IA³ adapters for specialized inference.
Embedding models with retrieval-quality measurements.
Cross-encoder rerankers with retrieval-stack receipts.
Curated and synthesized training/eval datasets.
Hosted demo spaces and interactive artifacts.
Open benchmarks and eval harnesses with publishable scores.
Every artifact is published alongside a field note. Read the running log of releases at /field-notes/.