aim-uofa / Self-Distillation-AgentRL
More Methods, Better Agents? A Reality Check on Self-Distillation in Agentic Reinforcement Learning
aim-uofa / Self-Distillation-AgentRL
titanwings / distilly
Distilly — Distill how they think into reusable Skills for any Agent or Bot. Formerly Colleague Skill（原同事 Skill）.
kangarooking / cangjie-skill
把书、长视频、播客等高价值内容蒸馏成可执行的 Agent Skills（Distill high-value content from books, long-form videos, podcasts, and more into executable Agent Skills）
PaddlePaddle / PaddleClas
A treasure chest for visual classification and recognition powered by PaddlePaddle
NVIDIA / Model-Optimizer
A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.
hao-ai-lab / FastVideo
A unified inference and post-training framework for accelerated video generation.
autodistill / autodistill
Images to inference with no labeling (use foundation models to train supervised models).
intel / neural-compressor
SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime
lightly-ai / lightly-train
All-in-one training for vision models (YOLO, ViTs, RT-DETR, DINOv3): pretraining, fine-tuning, distillation.
yoshitomo-matsubara / torchdistill
A coding-free framework built on PyTorch for reproducible deep learning studies. PyTorch Ecosystem. 🏆26 knowledge distillation methods presented at TPAMI, CVPR, ICLR, ECCV, NeurIPS, ICCV, AAAI, etc are implemented so far. 🎁 Trained models, training logs and configurations are available for ensuring the reproducibiliy and benchmark.
thu-ml / Causal-Forcing
[ICML 2026] Official codebase for "Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation" & Causal Forcing++
nolabs-ai / deepfabric
Generate High-Quality Synthetics, Train, Measure, and Evaluate in a Single Pipeline
worldbench / DiffusionOPSD
🔥 On-Policy Self-Distillation in Diffusion Models
mliu98 / awesome-human-distillation
A curated catalog of human distillliation agent skills
MeiGen-AI / OPSD-V
On-Policy Self-Distillation for Post-Training Few-Step Autoregressive Video Generators
nick7nlp / Awesome-LLM-On-Policy-Distillation
A curated collection of papers and resources on On-Policy Distillation for Large Language Models.
zengxiao-he / tessera
From teacher to tiles — a from-scratch LLM distillation & serving engine: custom Triton/CUDA kernels, FSDP distillation, paged-KV continuous batching, speculative decoding, a Rust gateway, a JAX oracle, and interpretability tooling.
AlibabaResearch / SparkDiffusion
SparkDiffusion: Accelerating DiT video generation by 265× via joint Sparsity, Distillation, Quantization, etc.
DaoyuanLi2816 / mini-verl
verl for a single consumer GPU. PPO, GRPO and on-policy distillation on NVIDIA GPUs.
zjwfufu / SMD
Simplifying and Improving On-Policy Distillation for Few-step Diffusion
Yzmblog / DMAD
DMAD: Distribution Matching as Adversarial Distillation for Fast Visual Generation
RH-RunningHub / MiniMax-H3-MultiGPU-Lightning
MiniMax-H3 multi-GPU inference acceleration: ~12x on 8x RTX 6000D (step distillation + SageAttention2 + Cache-DiT + torch.compile, TP2+Ulysses4 on sglang)
LiJiaxing0213 / DistillAlign
DistillAlign: Coordinating Mode Covering and Mode Seeking in Autoregressive Video Distillation
metaevo-ai / harness-zero
Harness-Zero: Harness Distillation via Agent-as-Harness
BytedTsinghua-SIA / Open-MOPD
Open-MOPD: Diagnosing and Fixing Capability Imbalance in Multi-Teacher On-Policy Distillation
fudan-generative-vision / WAM-Diff2
Hierarchical AR-to-Diffusion Distillation for Efficient Autonomous Driving VLA
dlgjr / FINAR-VL
Open-source financial multimodal LLM for document understanding, chart/table reasoning, numerical reasoning, and financial analysis, with SFT, dual-branch RL, MOPD distillation, and public benchmark evaluation.
ASLP-lab / CN-MultiDialect-ASR
Official repository of the paper "On-Policy Self-Distillation for Multi-Dialect ASR: Mastering Dialects, Retaining Mandarin".
sahilmahendrakar / paradee
An 8M-parameter English text-to-speech model distilled from Kokoro-82M. 9 MB, runs 18x real time on one CPU thread.
Thinking-Space / One-Shot-OPD
Rethinking On-Policy Distillation of Large Language Models II: One Training Example
ZJU-OmniAI / Spatial-Interactor
Learning Spatial Reasoning through Interaction with the Observable Physical World
uanu2002 / DAPD
DAPD: Dual-Anchored Policy Distillation
ZeamoxWang / pdmd
Official Training and Benchmarking Implementation of PDMD: Projected Distribution Matching Distillation for Video Diffusion Models
delaprada / Mask-Forcing
Official Implementation for Mask Forcing: Improving Autoregressive Video Diffusion Distillation via Dual-Noise Masking Rollout
lucidrains / light-loco-parkour
Implementation of Light Loco Parkour, Versatile Perceptive Whole-Body Locomotion via Multi-Skill Distillation, out of Light Origins Labs
nv-tlabs / cmd
Context-Matched Distillation: Teacher Causality for Autoregressive Video Distillation
ayutaz / sanoTTS-jp
559 K パラメータの日本語 TTS を ESP32-S3 で実時間合成。漢字かな交じり文の形態素解析・アクセント推定まで端末内で走る（M5Stack CoreS3 実機で確認）。推論は依存ゼロの C99、ブラウザ demo あり。arXiv:2608.21378 sanoTTS の日本語 clean-room 再実装。⚠️ コードは MIT ですが、配布モデルの重みは MIT ではありません（LICENSE-MODEL.md。出力に用途制限が伝播します）
tomerglick57 / Jevstiller
Distill a repeated Jev classification task into a local model, on the fly — same answers, your hardware.
bladedevoff / stuntd
Local proxy that learns your app's typed LLM decisions and answers them with a Laya head. Jev and OpenAI compatible.
InternLM / ResOPD
ResOPD: Tail Residualization for Sparse On-Policy Distillation. Core implementation built on verl.
alanhuangyoo / OpenJev
OpenJev: open-source, local alternative to Jev. System-One decision models: typed questions in, calibrated probabilities out; general decisions and browser-agent steps.
bingreeky / LOPD
Latent On-Policy Self-Distillation
VincentZhao2002 / OPOD
OPOD: On-Policy Omni Distillation - Code Release
DripNowhy / On-Policy-Self-Adaptation
Implementation of "Does On-Policy Distillation Really Distill? From Noisy Teacher to Self-Improvement"
LiXin97 / DN-MOPD
Beyond Teacher Assignment: Domain-Normalized Multi-Teacher On-Policy Distillation (DN-MOPD)
hufeng173 / kunpeng-skill
Kunpeng-Skill is a powerful multimodal distillation toolkit that transforms high-value insights from repositories, websites, UIs, videos, images, and documents into reusable methodologies, model-agnostic regeneration specifications, and more.
sanshengai / sansheng-distill
Claude Code skill: distill a book or video series into one interactive single-file HTML · 书籍/视频深度蒸馏
AllSpark-Research / D-3-MOPD
D^3-MOPD: DYNAMIC DOMAIN SCHEDULING FOR EFFICIENT MULTI-TEACHER DISTILLATION
davnords / poincar3
[Emergent Multi-View Geometry Through Self-Distillation]: 3D SSL through multi-view self-distillation, called Poincar3
jangviktor-web / tcm-distiller
中医思维蒸馏器TCM-Distiller是开源中医AI蒸馏流水线，不堆砌中医典籍素材，而是搭建标准化体系，把胡希恕、黄元御、吴鞠通等医家的著作、讲稿、医案，提炼成可被AI调用、还原真人辨证思路的专属思维技能。系统采用15维度蒸馏架构拆解医者心智模型、用药逻辑、剂量习惯与诊疗边界，搭配S1-S12十二重质控模块，依靠原文溯源、剂量勘误、盲评校验、索引审计等规则，解决普通AI训练杜撰医理、辨证混乱的问题。项目以棘轮迭代机制稳步优化还原度，仅用于中医研习与学术研究，内容不可作为临床诊疗方案，以数字化工程模式，打通传统中医与现代AI的传承通路。 TCM-Distiller is an open-source pipeline for TCM intellectual inheritance.
developeryj1992 / ai-trend-studio
Local-first AI trend radar and live-session distillation
wnby / travel-photo-soft-abstraction
A Codex skill for clean upper-photo/lower-abstraction travel artworks with source-readable soft visual distillation
zsxm1998 / awesome-perception-aware-rlvr
Unified reproductions, benchmarking & evaluation for perception-aware RLVR and OPD in VLMs. 19 methods · 30+ benchmarks · Official code of CGPO (ACM MM 2026 Oral).
LeapLabTHU / OPDVR
Official repository for OPDVR: On-policy Distillation with Verifiable Reward.
yusenthebot / stanford-ai-agent-skill
Unofficial distillation and executable skill for Stanford CS329A Self-Improving AI Agents
kissshhot / Awesome-OPD
A curated paper list on On-Policy Distillation (OPD) & On-Policy Self-Distillation (OPSD) for LLMs — student rollouts + teacher feedback, covering white-box/black-box methods, OPD-RL hybrids, agentic & multimodal applications, and industrial multi-teacher recipes.
jangviktor-web / zeng-shiqiang
曾仕强 AI Agent Skill，基于曾仕强书籍，电视节目及公开演讲等材料构建的大模型知识库 Skill 资源库。完整收录中国式管理全集 23 册、易经、周易、道德经、论语、财神文化、大易管理、三国智慧、家风家庭教育等人情世故与国学管理文本，全部整理为标准化 Markdown，完成实体书名与电子版别名映射、书目勘误、缺本标记。可直接用于 RAG 检索、LLM 微调、AI 蒸馏、Agent 调用、Obsidian 知识库构建、本地大模型私有 Skill 调用。覆盖修己安人、持经达变、领导统御、人际奥秘、亲子教养、易经中道思维、企业管理等场景，内置心智模型、口语范式与幻觉约束机制，规避 AI 编造输出，为 AI 提供准确的曾仕强国学思想输出素材，适合国学 RAG、私有大模型、中文 Agent
zencefilefendi / kadi
⚖️ Kadı — a calibrated judge in front of your frontier LLM. Routine decisions answered locally with a conformal-prediction guarantee (≤ α disagreement with the oracle); everything else escalated and learned from. Zero-dependency Python.