NexForge: Scaling Agent Capabilities through Requirement-Driven Task Synthesis for LLMs

Hugging Face Daily Papers Papers
requirement-driven-task-synthesis agent-training llm-post-training synthetic-data agent-capabilities scaling

Summary

NexForge is a requirement-driven framework that synthesizes diverse, executable agent tasks and expert trajectories for LLM post-training, outperforming prior methods and achieving state-of-the-art open-source agent performance on Terminal-Bench.

Scaling executable agent training data for LLM post-training is bottlenecked by substrate-bound methods that tie task generation to predefined tools, repositories, or skill graphs: expanding coverage requires manual substrate engineering, each new domain demands a bespoke pipeline, and the resulting task distributions often reflect substrate biases rather than real-world demand. We introduce NexForge, a requirement-driven framework that takes high-level capability requirements as input and synthesizes diverse, executable agent tasks and expert trajectories for SFT. NexForge first investigates real-world demand to construct representative scenarios and task profiles, then performs distribution-aware compilation to generate task directives. For each directive, NexForge automatically retrieves or constructs the required files, dependencies, and runtime configurations, and finally synthesizes expert rollouts and produces training trajectories. Without domain-specific infrastructure, NexForge produces 3.6K terminal and 2K office tasks, improving Qwen3.5-35B-A3B Base from 22.5\% to 52.0\% on Terminal-Bench 2.0 and from 813 to 1338 Elo on GDPval; scaling further to 43.2K terminal tasks yields 58.4\%, on par with Claude Opus 4.6 equipped with Claude Code. Scaled further, NexForge-synthesized data contributes to the training of Nex-N2, a family of publicly available agent models that lift Qwen3.5-35B-A3B to 75.3\% on Terminal-Bench 2.1 and to 1585 Elo on GDPval -- achieving state-of-the-art open-source performance and surpassing several frontier proprietary systems. Nex-N2 models are available at https://nex.sii.edu.cn/.
Original Article
View Cached Full Text

Cached at: 07/22/26, 02:40 AM

Paper page - NexForge: Scaling Agent Capabilities through Requirement-Driven Task Synthesis for LLMs

Source: https://huggingface.co/papers/2607.14186

Abstract

ScalingexecutableagenttrainingdataforLLMpost-trainingisbottleneckedbysubstrate-boundmethodsthattietaskgenerationtopredefinedtools,repositories,orskillgraphs:expandingcoveragerequiresmanualsubstrateengineering,eachnewdomaindemandsabespokepipeline,andtheresultingtaskdistributionsoftenreflectsubstratebiasesratherthanreal-worlddemand.WeintroduceNexForge,arequirement-drivenframeworkthattakeshigh-levelcapabilityrequirementsasinputandsynthesizesdiverse,executableagenttasksandexperttrajectoriesforSFT.NexForgefirstinvestigatesreal-worlddemandtoconstructrepresentativescenariosandtaskprofiles,thenperformsdistribution-awarecompilationtogeneratetaskdirectives.Foreachdirective,NexForgeautomaticallyretrievesorconstructstherequiredfiles,dependencies,andruntimeconfigurations,andfinallysynthesizesexpertrolloutsandproducestrainingtrajectories.Withoutdomain-specificinfrastructure,NexForgeproduces3.6Kterminaland2Kofficetasks,improvingQwen3.5-35B-A3BBasefrom22.5\%to52.0\%onTerminal-Bench2.0andfrom813to1338EloonGDPval;scalingfurtherto43.2Kterminaltasksyields58.4\%,onparwithClaudeOpus4.6equippedwithClaudeCode.Scaledfurther,NexForge-synthesizeddatacontributestothetrainingofNex-N2,afamilyofpubliclyavailableagentmodelsthatliftQwen3.5-35B-A3Bto75.3\%onTerminal-Bench2.1andto1585EloonGDPval--achievingstate-of-the-artopen-sourceperformanceandsurpassingseveralfrontierproprietarysystems.Nex-N2modelsareavailableathttps://nex.sii.edu.cn/.

View arXiv pageView PDFProject pageAdd to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.14186 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2607.14186 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.14186 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

CLI-Universe: Towards Verifiable Task Synthesis Engine for Terminal Agents

Hugging Face Daily Papers

CLI-Universe is a synthesis engine that generates verifiable terminal-agent tasks via multi-dimensional capability taxonomy and evidence-guided research, producing a distilled dataset of 6,000 trajectories. Fine-tuning Qwen3-32B on this dataset achieves 33.4% on Terminal-Bench 2.0, setting a new state-of-the-art for open-source models at or below 32B parameters.