[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tool-trl":3},{"tool":4,"categoryPool":37},{"id":5,"name":6,"slug":7,"description":8,"url":9,"githubUrl":10,"logoUrl":11,"openSource":12,"categories":13,"repo":18,"company":28,"articles":35,"createdAt":36},"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",true,[14],{"id":-1,"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},"huggingface\u002Ftrl",19066,2903,261,"Python","Apache-2.0",[],"2026-08-13T08:56:52Z",false,{"id":29,"name":30,"slug":31,"url":32,"bio":33,"githubOrg":34},"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,[38,62,86,133],{"id":39,"name":40,"slug":41,"description":42,"url":43,"githubUrl":43,"logoUrl":44,"openSource":12,"categories":45,"repo":47,"company":54,"articles":60,"createdAt":61},"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],{"id":-1,"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},"NVIDIA-NeMo\u002Flabs-molt",904,86,11,[],"2026-08-13T02:44:48Z",{"id":55,"name":56,"slug":57,"url":58,"bio":59,"githubOrg":56},"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":63,"name":64,"slug":65,"description":66,"url":67,"githubUrl":67,"logoUrl":68,"openSource":12,"categories":69,"repo":71,"company":78,"articles":84,"createdAt":85},"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",[70],{"id":-1,"name":15,"slug":16,"count":17},{"fullName":72,"url":67,"stars":73,"forks":74,"openIssues":75,"language":23,"license":24,"topics":76,"pushedAt":77,"archived":27},"ramp-public\u002Fportallib",113,5,3,[],"2026-07-27T21:38:51Z",{"id":79,"name":80,"slug":81,"url":82,"bio":-1,"githubOrg":83},"entity_01kzmf00gwe8krkm7xkvnn42dn","Ramp","ramp","https:\u002F\u002Framp.com\u002F","ramp-public",[],1785191314,{"id":87,"name":88,"slug":89,"description":90,"url":91,"githubUrl":92,"logoUrl":93,"openSource":12,"categories":94,"repo":99,"company":106,"articles":111,"createdAt":132},"entity_01kyajqefaesmb8qynd8f15097","Tinker","tinker","Tinker is a training API for researchers and developers.","https:\u002F\u002Fthinkingmachines.ai\u002Ftinker\u002F","https:\u002F\u002Fgithub.com\u002Fthinking-machines-lab\u002Ftinker","https:\u002F\u002Fr2.minima.ltd\u002Forg_01jakhtww5fk6b0jmxj0qfp9rf\u002Fapple-touch-icon-5d459972.png",[95,98],{"id":-1,"name":96,"slug":97,"count":17},"Frameworks & SDKs","frameworks-sdks",{"id":-1,"name":15,"slug":16,"count":17},{"fullName":100,"url":92,"stars":101,"forks":102,"openIssues":103,"language":23,"license":24,"topics":104,"pushedAt":105,"archived":27},"thinking-machines-lab\u002Ftinker",693,78,30,[],"2026-08-08T03:35:24Z",{"id":107,"name":108,"slug":109,"url":110,"bio":-1,"githubOrg":109},"entity_01kyn5g3kce4frj5e6rt2c3rpm","Thinking Machines Lab","thinking-machines-lab","https:\u002F\u002Fthinkingmachines.ai",[112,123],{"id":113,"name":114,"slug":115,"summary":116,"url":117,"kind":118,"platform":119,"author":108,"authorHandle":-1,"publisher":108,"publishedAt":120,"about":121,"writtenBy":122,"createdAt":-1},"entity_01kytdkp68e8jsra7tjvswdf63","Introducing Inkling-Small","introducing-inkling-small","An open-weights model that matches Inkling at a quarter of the size: multimodal, Mixture-of-Experts, with controllable reasoning effort. Fine-tune it on Tinker.","https:\u002F\u002Fthinkingmachines.ai\u002Fnews\u002Finkling-small\u002F","article","web","2026-07-30",[],[],{"id":124,"name":125,"slug":126,"summary":127,"url":128,"kind":118,"platform":119,"author":108,"authorHandle":-1,"publisher":108,"publishedAt":129,"about":130,"writtenBy":131,"createdAt":-1},"entity_01kyajeh43f6f93ar89zr8zsvg","Inkling: Our Open-Weights Model","inkling-our-open-weights-model","Our first open-weights model: multimodal, Mixture-of-Experts, with controllable reasoning effort. Available to fine-tune on Tinker.","https:\u002F\u002Fthinkingmachines.ai\u002Fnews\u002Fintroducing-inkling\u002F","2026-07-15",[],[],1784914098,{"id":5,"name":6,"slug":7,"description":8,"url":9,"githubUrl":10,"logoUrl":11,"openSource":12,"categories":134,"repo":136,"company":141,"articles":142,"createdAt":36},[135],{"id":-1,"name":15,"slug":16,"count":17},{"fullName":19,"url":10,"stars":137,"forks":138,"openIssues":22,"language":23,"license":24,"topics":139,"pushedAt":140,"archived":27},19064,2902,[],"2026-08-13T07:47:41Z",{"id":29,"name":30,"slug":31,"url":32,"bio":33,"githubOrg":34},[]]