OpenCodePapers

agents-on-bench-retail

Agents
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeAccuracyModelNameReleaseDate
Trajectory-Level Speculation with Semantic Verification for Tool-Using Agents✓ Link87.7SpecAgent2026-10-15
Progressive Agent Skill Generation via Reinforcement Learning✓ Link86.0Claude Sonnet 4.5 + Skill-α2026-08-03
Rufus-Air: An Open LLM Post-Training Recipe85.3Rufus-Air2026-09-24
PolicyGuide: From Guarding One Action to Guiding the Whole Workflow for Policy-Compliant LLM Agents✓ Link80.9PolicyGuide (GPT-5.4)2026-08-20
Kimi K2: Open Agentic Intelligence✓ Link70.6Kimi K2 Instruct2025-07-28
Towards General Agentic Intelligence via Environment Scaling✓ Link70.2AgentScaler-30B-A3B2025-09-16
Kolibri: A Sovereign European Model on the Pareto Frontier✓ Link69.9Kolibri2026-10-03
MUA-RL: Multi-turn User-interacting Agent Reinforcement Learning for agentic tool use67.3MUA-RL-32B2025-08-26
Unlocking Lossless Speedups in LLMs via Discrete Diffusion✓ Link67.1Uno (K2-Horizon-7B-Uno)2026-09-03
NVIDIA Nemotron 3: Efficient and Open Intelligence✓ Link62.83Nemotron 3 Super2025-12-24
CoVe: Training Interactive Tool-Use Agents via Constraint-Guided Verification59.4CoVe-4B2026-03-02
Predicting Privileged Information Effectiveness for On-Policy Self-Distillation✓ Link55.04PI-OPSD (Qwen3-4B-Instruct-2507)2026-10-15
LFM2.5-350M: No Size Left Behind17.84LFM2.5-350M2026-03-31
LFM2.5-230M: Built to Run Anywhere13.68LFM2.5-230M2026-06-25
Check out how to contribute new results. Then edit this file on GitLab or GitHub.