OpenCodePapers

agents-on-jobbench-main-65-tasks

Agents
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeScoreModelNameReleaseDate
JobBench Leaderboard✓ Link75.4Claude Opus 5.5 (OpenCode)2026-09-22
Muse Wonder: A Frontier Model For Personal & Professional AI Agents75.3Muse Wonder Full (max) (OpenCode)2026-10-12
JobBench Leaderboard73.1Claude Fable 5.1 (OpenCode)2026-09-01
Muse Wonder: A Frontier Model For Personal & Professional AI Agents70.5Muse Wonder (max) (OpenCode)2026-10-12
JobBench Leaderboard✓ Link66.1GPT-6 Astra (OpenCode)2026-09-03
JobBench Leaderboard✓ Link65.7Claude Opus 5 (OpenCode)2026-07-24
JobBench Leaderboard64.9Muse Spark 1.3 (OpenCode)2026-09-02
Introducing Muse Spark Open, The Most Capable US Open-Weight Model to Date✓ Link61.8Muse Spark Open2026-10-15
JobBench Leaderboard61.6Muse Spark 1.2 (OpenCode)2026-08-05
JobBench Leaderboard✓ Link61.4GLM-5.3 (OpenCode)2026-08-14
JobBench Leaderboard✓ Link57.4Claude Fable 5 (with fallbacks) (OpenCode)2026-06-09
Qwen3.8-Flash-Next✓ Link55.7Qwen3.8-Flash-Next2026-08-24
JobBench Leaderboard✓ Link54.7Muse Spark 1.1 (OpenCode)2026-07-09
JobBench Leaderboard✓ Link54.3Kimi K3 (OpenCode)2026-07-27
JobBench Leaderboard✓ Link52.7Qwen3.8-Max (OpenCode)2026-08-03
JobBench Leaderboard48.4Claude Opus 4.8 (OpenCode)2026-05-28
JobBench Leaderboard✓ Link45.4GPT-5.6 Sol (OpenCode)2026-07-09
JobBench Leaderboard44.5Claude Opus 4.7 (OpenCode)2026-04-16
JobBench Leaderboard✓ Link43.4GLM-5.2 (OpenCode)2026-06-16
JobBench Leaderboard38.3GPT-5.5 (OpenCode)2026-04-23
JobBench Leaderboard36.6Claude Sonnet 4.6 (OpenCode)2026-02-17
JobBench Leaderboard32.2GPT-5.4 (OpenCode)2026-03-05
JobBench Leaderboard31.5Gemini 3.5 Flash (OpenCode)2026-05-19
JobBench Leaderboard✓ Link31.1Inkling (OpenCode)2026-07-15
JobBench Leaderboard✓ Link27.2Nemotron 3 Ultra (OpenCode)2026-06-03
JobBench Leaderboard26.6GPT-5.2 (OpenCode)2025-12-11
JobBench Leaderboard20.7Claude Sonnet 4.5 (OpenCode)2025-09-29
JobBench Leaderboard15.9Gemini 3.1 Pro (OpenCode)2026-02-19