Back to Archive

AutoBench Agentic Run 1 - April 2026

The first AutoBench run to measure agentic performance of top LLMs

Past
Date
April 16, 2026
Version
2026-04-16
Models
31
New Models
22

Run data

Model
Average (All Topics)Adaptive ReplanningApi WorkflowDomain WorkflowError HandlingFailure RecoveryMulti Step OrchestrationParallel ExecutionParameter ComplexitySingle Tool CallTool Selection
21.55s (#1)26.34s (#2)27.95s (#3)30.02s (#3)21.89s (#2)21.69s (#1)15.67s (#1)29.06s (#2)18.66s (#4)10.02s (#1)14.23s (#2)
25.79s (#2)27.29s (#3)27.29s (#2)25.37s (#1)24.23s (#3)26.88s (#2)19.58s (#2)55.17s (#4)22.61s (#6)15.71s (#4)13.78s (#1)
30.73s (#3)34.73s (#4)51.20s (#8)34.91s (#5)37.86s (#5)30.83s (#3)29.59s (#4)43.74s (#3)13.65s (#2)14.12s (#3)16.67s (#5)
36.45s (#4)40.97s (#6)34.92s (#5)36.33s (#6)30.20s (#4)46.45s (#4)32.10s (#5)78.24s (#9)17.63s (#3)25.20s (#7)22.47s (#7)
41.25s (#5)22.59s (#1)18.28s (#1)26.93s (#2)17.24s (#1)243.06s (#26)20.71s (#3)26.23s (#1)8.45s (#1)11.09s (#2)17.96s (#6)
43.97s (#6)41.44s (#7)37.53s (#6)33.52s (#4)49.97s (#6)58.63s (#6)47.89s (#7)109.44s (#13)22.40s (#5)23.32s (#5)15.61s (#4)
54.19s (#7)45.97s (#8)69.68s (#11)43.21s (#7)53.61s (#7)57.12s (#5)57.55s (#9)72.84s (#7)49.42s (#10)63.73s (#16)28.82s (#8)
57.90s (#8)40.50s (#5)53.45s (#9)43.38s (#8)55.13s (#9)87.10s (#13)52.62s (#8)113.27s (#14)46.15s (#9)43.25s (#10)44.14s (#11)
63.36s (#9)51.60s (#9)33.31s (#4)73.13s (#12)66.80s (#11)76.08s (#9)43.41s (#6)91.52s (#11)56.49s (#11)126.51s (#23)14.77s (#3)
68.17s (#10)70.51s (#12)63.29s (#10)70.59s (#11)73.86s (#12)65.16s (#7)147.43s (#21)62.53s (#5)45.93s (#8)39.81s (#8)42.55s (#10)
76.10s (#11)64.54s (#11)73.80s (#12)67.02s (#10)63.99s (#10)82.16s (#10)118.40s (#16)76.50s (#8)72.38s (#15)68.62s (#18)73.61s (#20)
95.82s (#12)98.08s (#16)86.55s (#13)110.61s (#16)54.03s (#8)88.82s (#14)236.05s (#25)86.72s (#10)73.69s (#16)56.25s (#14)67.38s (#18)
97.33s (#13)80.04s (#13)197.75s (#25)100.16s (#15)80.42s (#13)86.53s (#12)110.94s (#15)150.53s (#19)69.72s (#14)44.96s (#13)52.25s (#13)
97.82s (#14)60.99s (#10)180.76s (#19)61.16s (#9)297.19s (#26)70.48s (#8)86.84s (#10)72.18s (#6)67.25s (#13)40.27s (#9)41.12s (#9)
109.38s (#15)99.64s (#17)113.88s (#17)180.36s (#21)163.17s (#17)85.53s (#11)90.04s (#12)114.39s (#15)58.42s (#12)90.51s (#22)97.86s (#23)
114.39s (#16)158.95s (#20)43.21s (#7)83.20s (#13)191.29s (#20)237.75s (#25)119.98s (#17)140.97s (#17)86.19s (#17)24.62s (#6)57.78s (#16)
134.63s (#17)129.62s (#18)110.14s (#15)95.46s (#14)303.95s (#27)131.78s (#16)89.62s (#11)212.00s (#23)160.59s (#25)63.75s (#17)49.37s (#12)
138.83s (#18)174.36s (#22)185.77s (#21)131.29s (#19)142.23s (#14)201.72s (#21)135.57s (#18)185.40s (#21)118.55s (#21)43.37s (#11)70.04s (#19)
147.14s (#19)154.34s (#19)189.27s (#23)278.34s (#27)153.42s (#15)207.92s (#22)104.01s (#13)141.66s (#18)134.93s (#23)43.71s (#12)63.82s (#17)
152.27s (#20)91.08s (#14)110.27s (#16)127.08s (#18)276.19s (#24)113.23s (#15)243.84s (#26)95.02s (#12)97.59s (#18)77.92s (#20)290.47s (#29)
156.44s (#21)188.98s (#24)107.32s (#14)112.58s (#17)185.41s (#18)259.09s (#28)250.29s (#27)125.74s (#16)221.95s (#28)57.34s (#15)55.67s (#15)
165.52s (#22)185.96s (#23)182.14s (#20)169.19s (#20)208.18s (#21)174.71s (#18)141.33s (#19)205.74s (#22)137.31s (#24)145.45s (#24)105.24s (#24)
174.48s (#23)334.16s (#29)122.39s (#18)189.85s (#22)158.96s (#16)191.31s (#19)104.83s (#14)274.55s (#25)124.98s (#22)162.15s (#25)81.63s (#21)
177.18s (#24)94.53s (#15)193.18s (#24)250.36s (#25)189.06s (#19)168.46s (#17)146.10s (#20)288.45s (#27)43.59s (#7)84.52s (#21)313.53s (#30)
197.75s (#25)167.48s (#21)251.43s (#29)285.74s (#28)312.55s (#29)233.09s (#24)168.08s (#22)295.67s (#28)104.81s (#19)72.48s (#19)86.12s (#22)
240.74s (#26)239.66s (#25)217.61s (#27)267.78s (#26)395.57s (#31)245.07s (#27)268.93s (#28)239.70s (#24)113.41s (#20)200.63s (#26)219.00s (#27)
240.88s (#27)333.09s (#28)217.50s (#26)303.03s (#31)239.12s (#23)302.55s (#31)198.36s (#23)179.33s (#20)267.11s (#29)313.20s (#29)55.52s (#14)
245.44s (#28)346.35s (#30)232.45s (#28)206.61s (#24)306.44s (#28)295.02s (#30)227.97s (#24)316.57s (#30)169.17s (#26)212.15s (#27)141.62s (#25)
262.21s (#29)254.31s (#26)188.87s (#22)196.32s (#23)291.05s (#25)196.74s (#20)301.91s (#29)327.64s (#31)320.11s (#30)320.85s (#30)224.27s (#28)
279.46s (#30)312.20s (#27)358.65s (#31)290.42s (#29)228.65s (#22)278.36s (#29)372.38s (#31)282.72s (#26)194.95s (#27)267.32s (#28)208.94s (#26)
329.39s (#31)357.31s (#31)304.61s (#30)300.90s (#30)395.21s (#30)208.72s (#23)333.41s (#30)313.73s (#29)396.42s (#31)330.23s (#31)353.36s (#31)
AutoBench Agentic Run 1 - April 2026 - AutoBench