{"id":4488,"date":"2026-07-01T15:46:23","date_gmt":"2026-07-01T15:46:23","guid":{"rendered":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/2026\/07\/01\/ornith-models-automate-agentic-coding-with-self-scaffolding\/"},"modified":"2026-07-01T15:46:23","modified_gmt":"2026-07-01T15:46:23","slug":"ornith-models-automate-agentic-coding-with-self-scaffolding","status":"publish","type":"post","link":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/2026\/07\/01\/ornith-models-automate-agentic-coding-with-self-scaffolding\/","title":{"rendered":"Ornith Models Automate Agentic Coding With Self-Scaffolding"},"content":{"rendered":"
\"\"<\/div>\n

\"\"<\/p>\n

Ornith, a new family of open source LLM models from the <\/span>DeepReinforce<\/span><\/a> research collective, takes a novel approach to executing coding and debugging tasks: It generates an architectural framework to give the user\u2019s harness a structured instruction set \u2013 a scaffold \u2013 to create an agent to complete the job.<\/span><\/p>\n

Available in a set of four variants, the <\/span>Ornith<\/span><\/a> family was trained to work comfortably with complex software repositories undertaking complicated long-horizon jobs. Sure, LLMs can do these tasks now \u2013 until the job gets too complex. Ornith\u2019s self-generated scaffolding ensures that it doesn\u2019t forget the plot along the way.<\/span><\/p>\n

\u201cThe model continuously improves not only its code generation abilities but also the orchestration strategy used to solve software engineering problems,\u201d wrote AI tutorial engineer Mehul Gupta, in an <\/span>introductory post<\/span><\/a>. <\/span><\/p>\n

Deep Reinforcement Expansion Pack<\/b><\/h3>\n

Ornith reads the user\u2019s instruction, but instead of executing it directly it builds a scaffold, a learnable object. The scaffold serves as a place where Ornith can design \u2013and refine \u2013 the architecture for the job.<\/span><\/p>\n

According to Gupta, the scaffold is where the LLM can detail the reasoning sequences, memory organization, debugging strategy, tool invocation order and execution planning. The user\u2019s harness then interprets the scaffold to generate an agent to execute the task.<\/span><\/p>\n

When the job is finished, the scaffold is deleted. When a new task comes up, Ornith builds a fresh scaffold to execute that job.<\/span><\/p>\n

\u201cBy jointly optimizing the scaffold and the resulting solution, the model can discover better search trajectories and generate higher-quality solutions,\u201d the researchers state in a<\/span> post<\/span><\/a>.<\/span><\/p>\n

Ornith builds the scaffolding from a set of rules developed in the model during training time. These models were built from an exhaustive self-learning process that used <\/span>deep reinforcement learning techniques<\/span><\/a> to computationally rotate through the possible ways of addressing an issue.<\/span><\/p>\n

Four Models<\/b><\/h3>\n

Ornith\u2019s <\/span> four variants<\/span><\/a> are: <\/span>9B Dense<\/b>, <\/span>31B Dense<\/b>, <\/span>35B MoE<\/b> and <\/span>397B MoE.<\/b> The \u201cDense\u201d models activate every parameter (as measured by the \u201cB\u201d in the name), whereas the MoE (Mixture of Experts) models activate only the parameters needed based on their relevance for the task, though they have additional reasoning tools for specialized functions.<\/span><\/p>\n

Each of the variants are built atop the open source Gemma 4 and Qwen 3.5, allowing the researchers to layer coding-specific deep RL rules over those models\u2019 inherent fluency in language and world knowledge. <\/span><\/p>\n

The dense models are best suited for running on local hardware. Ideal for a laptop, 9B Dense can write small scripts and execute various single-file cleanup tasks, whereas the 31B Dense requires a full workstation with up to 48GB of VRAM, but can internalize a full view of a complicated multi-file repository for tougher problems.<\/span><\/p>\n

The MoE variants are best run in the cloud. The 35B MoE is perhaps best suited for quick continuous integration patching and code review. The 397B MoE is the flagship model, a competitor to Opus 4.7, in the organization\u2019s estimation. This behemoth requires a cluster of GPUs to run smoothly, and can tackle the hardest coding problems.<\/span><\/p>\n

Killer Performance<\/b><\/h3>\n

With this diversity of models, Ornith\u2019s performance metrics are \u201cjust killing it all over the place,\u201d with impressive marks across small, middle and large LLM categories, <\/span>observed<\/span><\/a> the Hyderabad, Telangana-based <\/span>Data Science in Your Pocket<\/span><\/i><\/a> YouTube channel. This is \u201ca breakthrough \u2026 one of a kind model,\u201d they noted.<\/span><\/p>\n

In <\/span>company tests<\/span><\/a>, Ornith-1.0-397B outperformed Claude Opus 4.7 on Terminal-Bench 2.1, a benchmark for LLMs in terminal environments, scoring 77.5 to Claude\u2019s 70.3.<\/span><\/p>\n

Likewise, Ornith-1.0-35B significantly outperforms similar mid-sized models, including Qwen 3.5 (9 billion parameters) and Gemma 4 (12 billion parameters). It even rivaled the 31-billion-parameter Gemma 4 model. <\/span><\/p>\n

Read More<\/a><\/p>\n

\u200b<\/p>","protected":false},"excerpt":{"rendered":"

Ornith, a new family of open source LLM models from the DeepReinforce research collective, takes a novel approach to executing […]<\/p>\n","protected":false},"author":1,"featured_media":4489,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[5],"tags":[],"class_list":["post-4488","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-devops"],"_links":{"self":[{"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/posts\/4488","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/comments?post=4488"}],"version-history":[{"count":0,"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/posts\/4488\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/media\/4489"}],"wp:attachment":[{"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/media?parent=4488"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/categories?post=4488"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/rssfeedtelegrambot.bnaya.co.il\/index.php\/wp-json\/wp\/v2\/tags?post=4488"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}