Back to Archive

AutoBench Agentic Run 1 - April 2026

The first AutoBench run to measure agentic performance of top LLMs

Latest
Date
April 19, 2026
Version
2026-04-19
Models
32
New Models
1

Run data

Model
Average (All Topics)Adaptive ReplanningApi WorkflowDomain WorkflowError HandlingFailure RecoveryMulti Step OrchestrationParallel ExecutionParameter ComplexitySingle Tool CallTool Selection
9.27s (#1)11.02s (#2)11.39s (#3)12.33s (#2)9.96s (#2)9.45s (#1)9.74s (#1)12.16s (#2)5.79s (#2)4.09s (#1)5.78s (#2)
10.56s (#2)10.48s (#1)9.58s (#1)11.41s (#1)9.21s (#1)22.29s (#7)9.79s (#2)12.12s (#1)4.88s (#1)4.38s (#2)6.49s (#3)
12.38s (#3)13.71s (#3)11.14s (#2)13.89s (#4)15.14s (#5)13.88s (#3)14.92s (#5)19.48s (#5)6.38s (#3)5.69s (#3)5.48s (#1)
12.51s (#4)15.88s (#5)13.77s (#5)13.93s (#5)12.99s (#3)14.36s (#5)11.85s (#3)13.66s (#3)10.30s (#6)8.69s (#6)7.81s (#5)
13.14s (#5)15.39s (#4)14.56s (#6)13.67s (#3)13.48s (#4)13.44s (#2)14.05s (#4)19.84s (#6)8.31s (#5)8.10s (#5)8.42s (#6)
14.33s (#6)18.61s (#7)17.74s (#8)17.04s (#6)17.16s (#6)14.09s (#4)16.21s (#6)18.54s (#4)7.00s (#4)6.60s (#4)8.44s (#7)
18.03s (#7)18.21s (#6)12.63s (#4)20.54s (#8)25.83s (#7)20.13s (#6)22.44s (#8)20.26s (#7)14.20s (#9)14.53s (#10)6.51s (#4)
20.85s (#8)23.46s (#8)23.48s (#9)26.62s (#10)26.46s (#8)24.31s (#8)21.90s (#7)27.63s (#8)11.34s (#7)9.47s (#7)11.32s (#8)
23.17s (#9)25.97s (#9)15.68s (#7)18.29s (#7)29.88s (#10)31.61s (#12)36.07s (#12)29.29s (#9)13.92s (#8)12.31s (#8)17.37s (#12)
25.89s (#10)26.95s (#11)29.19s (#12)26.76s (#11)32.30s (#12)27.68s (#10)28.31s (#10)30.75s (#10)18.88s (#11)17.52s (#14)15.35s (#10)
25.92s (#11)26.14s (#10)28.82s (#11)25.29s (#9)27.89s (#9)26.67s (#9)27.39s (#9)34.50s (#13)23.25s (#14)16.62s (#12)20.46s (#13)
27.07s (#12)34.30s (#13)26.19s (#10)29.26s (#12)34.73s (#13)29.82s (#11)30.47s (#11)31.82s (#11)18.05s (#10)14.84s (#11)14.58s (#9)
33.09s (#13)32.96s (#12)46.91s (#20)36.79s (#14)35.31s (#14)31.72s (#13)36.91s (#13)35.66s (#14)35.66s (#22)19.61s (#17)23.81s (#16)
36.20s (#14)36.46s (#14)39.22s (#13)40.12s (#15)31.15s (#11)35.65s (#14)64.63s (#25)41.20s (#16)30.46s (#18)21.32s (#18)29.43s (#18)
37.87s (#15)42.32s (#16)54.46s (#22)46.71s (#19)42.76s (#15)38.19s (#16)38.72s (#15)44.29s (#17)29.02s (#17)17.29s (#13)21.89s (#14)
41.27s (#16)44.09s (#19)43.68s (#17)36.73s (#13)44.50s (#16)42.22s (#19)45.37s (#18)34.32s (#12)42.21s (#26)36.31s (#26)45.09s (#26)
42.87s (#17)43.70s (#18)43.95s (#18)49.20s (#20)60.11s (#21)48.82s (#21)37.08s (#14)49.67s (#18)32.33s (#20)22.81s (#19)30.26s (#21)
43.07s (#18)43.62s (#17)46.52s (#19)46.08s (#18)60.58s (#22)36.10s (#15)51.84s (#22)37.15s (#15)36.28s (#23)23.39s (#21)50.41s (#28)
43.55s (#19)50.45s (#20)40.41s (#14)43.28s (#16)66.69s (#23)39.12s (#17)48.02s (#20)52.61s (#20)36.52s (#24)23.05s (#20)22.97s (#15)
45.28s (#20)39.20s (#15)47.50s (#21)70.79s (#25)50.06s (#19)43.48s (#20)51.45s (#21)69.19s (#25)20.98s (#12)19.60s (#16)45.34s (#27)
46.43s (#21)57.86s (#22)40.92s (#15)59.86s (#22)58.99s (#20)41.77s (#18)43.74s (#17)58.85s (#22)22.29s (#13)31.20s (#24)40.97s (#24)
47.09s (#22)50.63s (#21)54.78s (#23)79.74s (#27)49.83s (#17)51.20s (#23)41.36s (#16)53.78s (#21)44.22s (#27)18.72s (#15)29.96s (#20)
53.60s (#23)73.68s (#25)59.48s (#24)43.43s (#17)49.87s (#18)81.78s (#27)74.00s (#27)62.39s (#23)25.23s (#16)26.89s (#23)29.63s (#19)
53.61s (#24)70.44s (#24)41.20s (#16)53.05s (#21)70.31s (#25)49.94s (#22)46.83s (#19)73.56s (#27)46.91s (#28)40.05s (#28)31.13s (#22)
56.85s (#25)84.84s (#27)69.59s (#28)61.18s (#23)79.18s (#26)69.15s (#24)58.65s (#23)81.88s (#28)23.93s (#15)12.66s (#9)16.30s (#11)
60.30s (#26)66.59s (#23)67.19s (#26)72.35s (#26)67.94s (#24)71.40s (#25)61.28s (#24)87.51s (#30)31.59s (#19)25.73s (#22)36.13s (#23)
69.40s (#27)100.80s (#28)63.74s (#25)82.07s (#29)79.34s (#27)88.14s (#29)83.45s (#28)69.54s (#26)33.77s (#21)37.39s (#27)44.53s (#25)
75.15s (#28)76.03s (#26)77.74s (#29)102.61s (#30)108.86s (#29)97.95s (#30)73.68s (#26)66.32s (#24)38.84s (#25)35.18s (#25)53.12s (#29)
86.91s (#29)153.36s (#31)90.63s (#30)81.13s (#28)90.32s (#28)101.93s (#31)103.88s (#30)49.85s (#19)91.57s (#31)78.56s (#30)26.54s (#17)
93.34s (#30)128.94s (#30)68.28s (#27)67.87s (#24)145.88s (#32)77.75s (#26)140.74s (#32)82.56s (#29)78.78s (#30)80.94s (#31)78.70s (#31)
105.95s (#31)128.85s (#29)126.32s (#32)119.78s (#32)117.21s (#30)102.23s (#32)140.19s (#31)127.50s (#32)78.06s (#29)52.46s (#29)67.39s (#30)
131.01s (#32)181.33s (#32)117.59s (#31)108.80s (#31)140.56s (#31)86.83s (#28)96.52s (#29)117.26s (#31)196.75s (#32)113.17s (#32)174.45s (#32)
AutoBench Agentic Run 1 - April 2026 - AutoBench