[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f5G4g7pZdZGTOwhopQ8xVmdMRAuJpjvMBhyrGXOF0GIo":3},{"tool":4,"categoryPool":31},{"id":5,"name":6,"slug":7,"description":8,"url":9,"githubUrl":10,"logoUrl":11,"openSource":12,"categories":13,"repo":18,"articles":29,"createdAt":30},"entity_01m0bh5ad4fztszqt82mqfptv3","Miles","miles","Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.","https:\u002F\u002Fmiles.radixark.com\u002F","https:\u002F\u002Fgithub.com\u002Fradixark\u002Fmiles","https:\u002F\u002Fr2.minima.ltd\u002Forg_01jakhtww5fk6b0jmxj0qfp9rf\u002Fapple-touch-icon-e1f83c20.png",true,[14],{"name":15,"slug":16,"count":17},"Post-Training","post-training",0,{"fullName":19,"url":10,"stars":20,"forks":21,"openIssues":22,"language":23,"license":24,"topics":25,"pushedAt":26,"archived":27,"licenseSpdx":24,"licenseOsi":12,"licenseVerifiedAt":28},"radixark\u002Fmiles",2091,369,898,"Python","Apache-2.0",[],"2026-08-19T00:29:38Z",false,"2026-08-18T22:51:39.779Z",[],1787093494,[32,38,63,87],{"id":5,"name":6,"slug":7,"description":8,"url":9,"githubUrl":10,"logoUrl":11,"openSource":12,"categories":33,"repo":35,"articles":37,"createdAt":30},[34],{"name":15,"slug":16,"count":17},{"fullName":19,"url":10,"stars":20,"forks":21,"openIssues":22,"language":23,"license":24,"topics":36,"pushedAt":26,"archived":27,"licenseSpdx":24,"licenseOsi":12,"licenseVerifiedAt":28},[],[],{"id":39,"name":40,"slug":41,"description":42,"url":43,"githubUrl":43,"logoUrl":44,"openSource":12,"categories":45,"repo":47,"company":55,"articles":61,"createdAt":62},"entity_01kyjhhkh2eh0v8dwkexzpmgmj","Molt","molt","An agentic-first RL framework for research. Ray · vLLM · NVIDIA AutoModel — the smallest PyTorch-native stack for 1T-class fully-async, multimodal, multi-turn agentic RL.","https:\u002F\u002Fgithub.com\u002FNVIDIA-NeMo\u002Flabs-molt","https:\u002F\u002Fr2.minima.ltd\u002Forg_01jakhtww5fk6b0jmxj0qfp9rf\u002FNVIDIA-NeMo-dd56beca.png",[46],{"name":15,"slug":16,"count":17},{"fullName":48,"url":43,"stars":49,"forks":50,"openIssues":51,"language":23,"license":24,"topics":52,"pushedAt":53,"archived":27,"licenseSpdx":24,"licenseOsi":12,"licenseVerifiedAt":54},"NVIDIA-NeMo\u002Flabs-molt",917,87,12,[],"2026-08-16T23:53:01Z","2026-08-09",{"id":56,"name":57,"slug":58,"url":59,"bio":60,"githubOrg":57},"entity_01kzrxjf0ze8h9epsbfe0560wm","NVIDIA","nvidia","https:\u002F\u002Fwww.nvidia.com","NVIDIA invents the GPU and drives advances in AI, HPC, gaming, creative design, autonomous vehicles, and robotics.",[],1785181294,{"id":64,"name":65,"slug":66,"description":67,"url":68,"githubUrl":68,"logoUrl":69,"openSource":12,"categories":70,"repo":72,"company":79,"articles":85,"createdAt":86},"entity_01kyjv3cm1fzvt3avvgastqaax","PorTAL","portal","PorTAL generates portable task specific LoRA adapters that can efficiently transfer across language models.","https:\u002F\u002Fgithub.com\u002Framp-public\u002Fportallib","https:\u002F\u002Fr2.minima.ltd\u002Forg_01jakhtww5fk6b0jmxj0qfp9rf\u002Framp-public-076e634e.png",[71],{"name":15,"slug":16,"count":17},{"fullName":73,"url":68,"stars":74,"forks":75,"openIssues":76,"language":23,"license":24,"topics":77,"pushedAt":78,"archived":27,"licenseSpdx":24,"licenseOsi":12,"licenseVerifiedAt":54},"ramp-public\u002Fportallib",113,6,3,[],"2026-07-27T21:38:51Z",{"id":80,"name":81,"slug":82,"url":83,"githubOrg":84},"entity_01kzmf00gwe8krkm7xkvnn42dn","Ramp","ramp","https:\u002F\u002Framp.com\u002F","ramp-public",[],1785191314,{"id":88,"name":89,"slug":90,"description":91,"url":92,"githubUrl":93,"logoUrl":94,"openSource":12,"categories":95,"repo":97,"company":104,"articles":111,"createdAt":112},"entity_01kyaxa5shecrr448eb44dg2mm","TRL","trl","TRL is a cutting-edge library designed for post-training foundation models using advanced techniques like Supervised Fine-Tuning (SFT), Group Relative Policy Optimization (GRPO), and Direct Preference Optimization (DPO). Built on top of the Transformers ecosystem, TRL supports a variety of model architectures and modalities, and can be scaled-up across various hardware setups.","https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftrl","https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Ftrl","https:\u002F\u002Fr2.minima.ltd\u002Forg_01jakhtww5fk6b0jmxj0qfp9rf\u002Fhuggingface-c9877a9c.png",[96],{"name":15,"slug":16,"count":17},{"fullName":98,"url":93,"stars":99,"forks":100,"openIssues":101,"language":23,"license":24,"topics":102,"pushedAt":103,"archived":27,"licenseSpdx":24,"licenseOsi":12,"licenseVerifiedAt":54},"huggingface\u002Ftrl",19103,2914,268,[],"2026-08-18T23:24:38Z",{"id":105,"name":106,"slug":107,"url":108,"bio":109,"githubOrg":110},"entity_01kynhneqheczterdgx1637yhz","Hugging Face","hugging-face","https:\u002F\u002Fhuggingface.co\u002F","We’re on a journey to advance and democratize artificial intelligence through open source and open science.","huggingface",[],1784925198]