Spaces:

---

ServiceNow

/

browsergym-leaderboard

like 26

Running

Adds MiniWoB leaderboard results for zip-glm47-sy.

Changes

• Added results/zip-glm47-sy/README.md with agent details:
  • Zip browser-agent runtime
  • z-ai/glm-4.7:nitro via OpenRouter
  • unified-mode BrowserGym/AgentLab setup
  • no fine-tuning
  • benchmark command/configuration
• Added results/zip-glm47-sy/miniwob.json with official leaderboard-format MiniWoB result:
  • Score: 86.4
  • Std err: 3.066
  • Study ID: 2026-05-26_15-41-13
  • Benchmark: MiniWoB
  • Original result, reproducible, not benchmark-specific/tuned