AgBench: Agentic AI Benchmarks for Personal AI Devices
cs.AI, cs.PF
Submitted: 2026-09-29
Updated: 2026-09-29
Comments: 15 pages, 12 figures, including supplementary material
Code: https://github.com/ggml-org/llama.cpp
Project page: https://www.amd.com/en/products/processors/consumer/agent-computers.html
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Agent S: An Open Agentic Framework that Uses Computers Like a Human
- PalmClaw: A Native On-Device Agent Framework for Mobile Phones
- TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents
- DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents
- Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
- When Cloud Agents Meet Device Agents: Lessons from Hybrid Multi-Agent Systems
- Agentic Much? Adoption of Coding Agents on GitHub
- The OpenHands Software Agent SDK: A Composable and Extensible Foundation for Production Agents
- UFO2: The Desktop AgentOS
- Hera: Learning Long-Horizon Coordination for Device-Cloud Collaborative LLM Agents
- Mobile-Agent-RAG: Driving Smart Multi-Agent Coordination with Contextual Knowledge Empowerment for Long-Horizon Mobile Automation
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems