Salesforce Koa: An Enterprise Language Model for Agentic Tool Use

• URL: http://arxiv.org/abs/2609.15066v1
• Date: Mon, 14 Sep 2026 05:30:28 GMT
• Title: Salesforce Koa: An Enterprise Language Model for Agentic Tool Use
• Abstract summary: We present Salesforce Koa, an enterprise language model built by post-training the open-weight Nemotron-3-Super-120B foundation model.<n>It is trained on public and synthetically generated data, with no customer data, to improve tool use and agentic capabilities.
• Score: 42.00565879954964
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We present Salesforce Koa, an enterprise language model built by post-training the open-weight Nemotron-3-Super-120B foundation model with reinforcement learning using Group Relative Policy Optimization (GRPO). Salesforce Koa is trained on public and synthetically generated data, with no customer data, to improve tool use and agentic capabilities while preserving strong general-purpose performance. Its distinctive component is a simulation-to-reward pipeline that expands workflow specifications into persona-conditioned multi-turn tasks with task-resolution rewards grounded in successful tool use for data-dependent requests. For enterprise domains, these specifications are written in Agent Script, Salesforce's declarative language for building Agentforce agents; for public tool-use domains, we synthesize the workflow structure directly. The same simulation and grounded-reward machinery drives GRPO across both. Across public tool-use, agentic-reasoning, and enterprise Customer Relationship Management (CRM) benchmarks, Salesforce Koa improves over its open-weight base, with the clearest gains on multi-turn tool use, and surpasses a strong proprietary baseline while remaining below the strongest frontier models. These results show that specification-driven reinforcement learning is a practical path to specializing open-weight foundation models for enterprise agentic tasks.

Related papers

• AutoB2G: A Large Language Model-Driven Agentic Framework For Automated Building-Grid Co-Simulation [6.259798608712568]
  AutoB2G is an automated building-grid co-simulation framework based on natural-language task descriptions.<n>It can effectively enable automated simulator implementations, coordinating B2G interactions to improve grid-side performance metrics.
  arXiv  Detail & Related papers  (2026-03-27T01:44:08Z)
• Let It Flow: Agentic Crafting on Rock and Roll, Building the ROME Model within an Open Agentic Learning Ecosystem [90.17610617854247]
  We introduce the Agentic Learning Ecosystem (ALE), a foundational infrastructure that optimize the production pipeline for agentic model.<n>ALE consists of three components: ROLL, a post-training framework for weight optimization; ROCK, a sandbox environment manager for trajectory generation; and iFlow CLI, an agent framework for efficient context engineering.<n>We release ROME, an open-source agent grounded by ALE and trained on over one million trajectories.
  arXiv  Detail & Related papers  (2025-12-31T14:03:39Z)
• Instruction-Tuning Open-Weight Language Models for BPMN Model Generation [0.0]
  We investigate whether open-weight large language models, adapted via instruction tuning, can generate high-quality BPMN process models.<n>We introduce InstruBPM, a reproducible approach that prepares paired text-diagram data and instruction tunes an open source large language model.<n>We compare the tuned model with untuned open-weight baselines and strong proprietary models under consistent prompting regimes.
  arXiv  Detail & Related papers  (2025-12-12T22:07:51Z)
• Klear-AgentForge: Forging Agentic Intelligence through Posttraining Scaling [46.593200463657645]
  We present a comprehensive and fully open-source pipeline for training a high-performance agentic model, named Klear-Qwen3-AgentForge.<n>We design effective supervised fine-tuning (SFT) with synthetic data followed by multi-turn reinforcement learning (RL) to unlock the potential for multiple diverse agentic tasks.
  arXiv  Detail & Related papers  (2025-11-08T09:47:27Z)
• SCUBA: Salesforce Computer Use Benchmark [63.66753028386581]
  SCUBA is a benchmark designed to evaluate computer-use agents on customer relationship management ( CRM) within the Salesforce platform.<n> SCUBA contains 300 task instances derived from real user interviews, spanning three primary personas, platform administrators, sales representatives, and service agents.<n>We benchmark a diverse set of agents under both zero-shot and demonstration-augmented settings.
  arXiv  Detail & Related papers  (2025-09-30T16:48:49Z)
• OpenCUA: Open Foundations for Computer-Use Agents [74.61449905487565]
  Vision-language models have demonstrated impressive capabilities as computer-use agents (CUAs)<n>We propose OpenCUA, a comprehensive open-source framework for scaling CUA data and foundation models.<n>Our end-to-end agent models demonstrate strong performance across CUA benchmarks.
  arXiv  Detail & Related papers  (2025-08-12T17:52:32Z)
• Agent WARPP: Workflow Adherence via Runtime Parallel Personalization [0.0]
  Large language models (LLMs) are increasingly applied in task-oriented dialogue (TOD) systems.<n>We present Adherence via Parallel Personalization, or WARPP, a training-free, modular framework that combines multi-agent runtime with orchestration.<n>By dynamically pruning conditional branches based on user attributes, the framework reduces reasoning overhead and narrows tool selection at runtime.
  arXiv  Detail & Related papers  (2025-07-23T23:27:49Z)
• Command A: An Enterprise-Ready Large Language Model [180.18356391290172]
  Command A is an agent-optimised and multilingual-capable model.<n>It offers best-in-class Retrieval Augmented Generation capabilities.
  arXiv  Detail & Related papers  (2025-04-01T12:08:07Z)
• MIMIR: A Streamlined Platform for Personalized Agent Tuning in Domain Expertise [49.83486066403154]
  textscMimir is a streamlined platform offering a customizable pipeline for personalized agent tuning. textscMimir supports the generation of general instruction-tuning datasets from the same input. textscMimir integrates these features into a cohesive end-to-end platform, facilitating everything from the uploading of personalized files to one-click agent fine-tuning.
  arXiv  Detail & Related papers  (2024-04-03T23:42:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.