Muse Spark 1.1 Debuts at 6th on APEX-Agents
giffmana · x · 2026-07-14
Muse Spark 1.1 has made its debut on APEX-Agents, ranking 6th.
- Pass@1: 37.1%, Mean criteria passed: 52.0%.
- Officials noted that in about 10% of the tasks, the model failed before completing the trajectory, resulting in an automatic score of 0 for these tasks.
- If trajectories impacted by Meta's aggressive content filtering are excluded, Pass@1 would rise to 41.4%, potentially ranking it ahead of GPT-5.6 Sol.
- This benchmark is designed for long-horizon tasks based on real-world banking, legal, and consulting work.
More from Models
- Claim says Kimi was distilled from Fable, sparking a model-attribution jab — cephaloform · 2026-07-22
- Gary Marcus says LLM math skills are like knowing only a car’s engine size — GaryMarcus · 2026-07-22
- OpenAI’s Codex + GPT-5.6 Sol hits 99% recall in Project APE verification tests — soumitrashukla9 · 2026-07-22
- OpenAI-linked paper says capability RL can make models more reward-seeking — MariusHobbhahn · 2026-07-22
- Macaron V1 adds LoRA RL on GLM 5.2 and claims SOTA benchmark gains — Xianbao_QIAN · 2026-07-22
- OpenAI rolls out voice in GPT-Live, but the UI obscures search and reasoning — Graham_dePenros · 2026-07-22