--- title: Supercharging LLMs: Scalable RL with torchforge and... description: Meta's PyTorch team open-sourced torchforge, a PyTorch-native RL library designed to simplify large-scale post-training of LLMs. In collaboration with Stanford... canonical: https://daily.dev/posts/supercharging-llms-scalable-rl-with-torchforge-and-weaver-pytorch-oqim6pm4y twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Supercharging LLMs: Scalable RL with torchforge and Weaver – PyTorch | daily.dev og:description: Meta's PyTorch team open-sourced torchforge, a PyTorch-native RL library designed to simplify large-scale post-training of LLMs. In collaboration with Stanford... og:url: https://daily.dev/posts/supercharging-llms-scalable-rl-with-torchforge-and-weaver-pytorch-oqim6pm4y og:image: https://api.daily.dev/og/posts/oqIM6pm4Y.png og:image:alt: Supercharging LLMs: Scalable RL with torchforge and Weaver – PyTorch og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Supercharging LLMs: Scalable RL with torchforge and Weaver – PyTorch **[PyTorch](https://daily.dev/sources/pytorch)** · 10 min read · 1 upvotes · 0 comments ## Summary Meta's PyTorch team open-sourced torchforge, a PyTorch-native RL library designed to simplify large-scale post-training of LLMs. In collaboration with Stanford and CoreWeave, they demonstrated scalable reinforcement learning on a 512-GPU cluster using GRPO with Weaver as a verifier. Weaver aggregates multiple weak verifiers to provide reliable reward signals without human annotations, achieving 44-65% of the performance gap between single reward models and fully annotated training across MATH-500, GPQA, and MMLU Pro benchmarks. The stack combines torchforge for RL primitives, Weaver for verification, and Monarch for distributed coordination, enabling researchers to iterate on RL algorithms without rebuilding infrastructure. Results show 4x faster iteration, >90% job completion rate, and >65% GPU utilization at scale. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [Introducing torchforge – a PyTorch native library for scalable RL post-training and agentic development – PyTorch](https://daily.dev/posts/introducing-torchforge-a-pytorch-native-library-for-scalable-rl-post-training-and-agentic-developm-vlg3fasxe) · PyTorch · 1 upvotes · 0 comments - [GRL: Turning verifiable games into a post-training suite for LLM agents with Tunix on TPUs](https://daily.dev/posts/grl-turning-verifiable-games-into-a-post-training-suite-for-llm-agents-with-tunix-on-tpus-atklkqbnh) · Google Open Source Blog · 2 upvotes · 0 comments --- Tags: [#machine-learning](https://daily.dev/tags/machine-learning), [#llm](https://daily.dev/tags/llm), [#pytorch](https://daily.dev/tags/pytorch), [#distributed-systems](https://daily.dev/tags/distributed-systems), [#reinforcement-learning](https://daily.dev/tags/reinforcement-learning) [View this post on daily.dev](https://daily.dev/posts/supercharging-llms-scalable-rl-with-torchforge-and-weaver-pytorch-oqim6pm4y) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Supercharging LLMs: Scalable RL with torchforge and Weaver – PyTorch","url":"https://daily.dev/posts/supercharging-llms-scalable-rl-with-torchforge-and-weaver-pytorch-oqim6pm4y","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/supercharging-llms-scalable-rl-with-torchforge-and-weaver-pytorch-oqim6pm4y"},"datePublished":"2026-01-09T20:38:39.836Z","dateModified":"2026-01-09T20:39:05.611Z","description":"Meta's PyTorch team open-sourced torchforge, a PyTorch-native RL library designed to simplify large-scale post-training of LLMs. In collaboration with Stanford...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/6c6e8ee318693c4de6f5f85f45160b68?_a=AQAEulh","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/6c6e8ee318693c4de6f5f85f45160b68?_a=AQAEulh","isAccessibleForFree":true,"articleSection":"PyTorch","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"PyTorch","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/5fdab3d75a834a97a52d433d9c7e5ff9","url":"https://daily.dev/sources/pytorch"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/supercharging-llms-scalable-rl-with-torchforge-and-weaver-pytorch-oqim6pm4y","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"machine-learning,llm,pytorch,distributed-systems,reinforcement-learning","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"PyTorch","item":"https://daily.dev/sources/pytorch"},{"@type":"ListItem","position":3,"name":"Supercharging LLMs: Scalable RL with torchforge and Weaver – PyTorch"}]} ```