All Dev Resources
3093 resources total
Kage
★ 90Shadows any website to a single binary for offline viewing.
Bambuddy
★ 90Open-source, self-hosted, cloud-free central management for Bambu Lab printers.
PixelRAG
★ 90A system that renders web pages as screenshots for RAG, bypassing text parsing errors and improving retrieval accuracy.
Pyodide
★ 90A Python distribution for the browser and Node.js based on WebAssembly.
CAP-TTA
★ 88A test-time adaptation framework for fast, stable debiasing in LLM narrative generation.
MDL-inspired framework for layer-adaptive LLM optimization and pruning.
Jasper
★ 88A GPU-native ANNS system with high query throughput and updatability using RaBitQ quantization.
AgenticRepair
★ 88An agentic vulnerability repair framework using multi-faceted program context engineering.
RAID
★ 88Robust AI-generated image detection using bit-reversed images and convolutional classifiers.
Harvest Benchmark
★ 88A benchmark for AI coding agents curated from real developer prompts and production tests.
Pretrained models and fine-tuning datasets for ARC-Encoder.
DataFlow-Harness
★ 88Open-source framework for guiding LLM agents to build structured, visual data-processing workflows.
Ratchet
★ 88A framework for self-evolving LLM agents to curate and retire their own skill libraries.
Pegasus
★ 88Low-resource framework for translating human demonstration videos into robot-learnable data.
An open-source Python toolkit for fingerprinting AI model weights to determine their lineage.
REPREC
★ 88Parameter-efficient recommendation system framework using lightweight user representation alignment.
OrchBench
★ 88Simulation-based benchmark for evaluating multi-agent orchestration plans in isolation.
HANDBOOK.md
★ 88A benchmark for long-context agentic instruction following using enterprise-style policy documents.
ContractHIL-HLS
★ 88Multi-agent workflow for contract-aligned HLS design with hardware-in-the-loop feedback.
SpecPrefetch
★ 88A parameter-efficient prefetching framework for offloaded MoE inference to reduce loading latency.