- #17max of 4443.8%
Leader: Gemini 3.7 Flash · 60.0%
- #64max of 2241336
- #83xhigh of 2241237
- #89high of 2241186
- #102medium of 2241099
- #135none of 224903
- #154low of 224741
Leader: Sonnet 5.5 · 1823
- #26max of 501.9%
Leader: Opus 5.5 · 15.5%
agentic
AutomationBench (Zapier)
115 results- #36max of 11520.7%
- #55high of 11514.5%
- #63xhigh of 11512.6%
- #74medium of 1159.4%
- #109low of 1151.2%
- #113none of 1150.3%
Leader: Gemini 4 · 51.3%
agentic
AutomationBench-AA
219 results- #66max of 21953.2%
- #79high of 21947.8%
- #81xhigh of 21947.8%
- #93medium of 21940.5%
- #145low of 21912.0%
- #148none of 2199.1%
Leader: Gemini 4 · 77.5%
- #24max of 2761.5%
Leader: Sonnet 5.5 · 81.1%
agentic
Harvey's Legal Agent Benchmark
79 results- #43max of 792.9%
Leader: Muse Spark 1.2 · 25.4%
agentic
Legal Research Bench
78 results- #46max of 7830.3%
Leader: Muse Spark 1.3 · 55.3%
- #46max of 6919.9%
Leader: Fable 5.1 · 49.2%
agentic
Terminal-Bench 4.0 (Vals)
47 results- #31max of 4713.6%
Leader: Opus 5.5 · 65.2%
agentic
Terminal-Bench Science (Vals)
40 results- #25max of 404.3%
Leader: GPT-6 Astra · 62.9%
- #11xhigh of 177.9%
Leader: GPT-6 Astra · 42.2%
coding
Arena Image-to-WebDev
58 results- #24max of 581570
Leader: Opus 5.5 · 1749
- #33max of 1221581
Leader: Opus 5.5 · 1813
- #27max of 7742.6%
Leader: Sonnet 5.5 · 69.8%
coding
FrontierCode 1.1 Extended
130 results- #56max of 13056.1%
- #79xhigh of 13051.4%
- #82high of 13051.3%
- #90medium of 13050.1%
- #112low of 13039.1%
Leader: Opus 5.5 · 65.3%
coding
FrontierCode 1.1 Main
130 results- #56max of 13042.4%
- #80high of 13037.3%
- #81xhigh of 13037.1%
- #87medium of 13035.5%
- #108low of 13025.7%
Leader: Opus 5.5 · 54.6%
- #21max of 4455.6%
Leader: Gemini 4 · 100.0%
coding
ProgramBench v1 Almost Resolved
62 results- #33max of 628.0%
Leader: Opus 5.5 · 65.0%
coding
ProgramBench v1 Fully Resolved
62 results- #25max of 620.5%
Leader: Opus 5.5 · 18.5%
coding
ProgramBench v1 Raw Pass Rate
62 results- #29max of 6266.2%
Leader: Opus 5.5 · 87.0%
- #58max of 22654.6%
- #88xhigh of 22651.7%
- #97medium of 22650.9%
- #104high of 22650.3%
- #130low of 22646.9%
- #151none of 22643.1%
Leader: Opus 5.5 · 66.9%
coding
Terminal-Bench 2.1 (Vals)
74 results- #23max of 7473.0%
Leader: Opus 5.5 · 87.6%
coding
Terminal-Bench Science-AA
46 results- #29max of 468.6%
Leader: GPT-6 Astra · 63.3%
coding
Terminal-Bench v4-AA
216 results- #77max of 21612.6%
- #90xhigh of 2168.1%
- #101high of 2164.5%
- #109medium of 2162.5%
- #122none of 2161.5%
- #158low of 2160.0%
Leader: Sonnet 5.5 · 63.6%
coding
Vibe Code Bench v1.1
105 results- #22max of 10581.6%
Leader: Sonnet 5.5 · 92.4%
composite
AA Intelligence Index v4.3.2
610 results- #75max of 61038.1
- #86xhigh of 61034.6
- #104high of 61032.9
- #121medium of 61029.9
- #206low of 61021.5
- #246none of 61018.5
Leader: Opus 5.5 · 57.6
- #76max of 2201443
Leader: Gemini 4 · 1525
composite
Arena Text — Multi-Turn
220 results- #85max of 2201444
Leader: Gemini 4 · 1553
- #28max of 4751.2%
Leader: Gemini 4 · 68.9%
cost
AA Intelligence Index Cost per Task
177 results- #1low of 177$0.00
- #7none of 177$0.01
- #12medium of 177$0.02
- #18high of 177$0.03
- #20xhigh of 177$0.04
- #30max of 177$0.07
Leader: this model · $0.00
cost
Cost to Run AA Intelligence Index
177 results- #1low of 177$10.63
- #4none of 177$19.16
- #10medium of 177$31.17
- #17high of 177$47.85
- #22xhigh of 177$66.81
- #30max of 177$122
Leader: this model · $10.63
- #24high of 341386
Leader: Opus 5.5 · 2079
creative
VoxelBench (text)
55 results- #16max of 551800
Leader: GPT-6 Astra · 2732
economics
Excel Modeling Benchmark
74 results- #15max of 7468.5%
Leader: Fable 5.1 · 76.7%
economics
Finance Agent v2
78 results- #43max of 7849.9%
Leader: Gemini 4 · 65.4%
- #47max of 21022.8%
- #70xhigh of 21018.6%
- #74high of 21018.2%
- #99medium of 21013.8%
- #135low of 2109.0%
- #153none of 2105.8%
Leader: Muse Wonder Full (Max) · 32.6%
- #65max of 2881430
- #80xhigh of 2881364
- #89high of 2881340
- #103medium of 2881262
- #148none of 2881045
- #153low of 2881033
Leader: Opus 5.5 · 1866
knowledge
AA-Omniscience Accuracy
526 results- #93xhigh of 52644.2%
- #94max of 52643.8%
- #99medium of 52643.1%
- #103high of 52642.8%
- #117low of 52640.8%
- #167none of 52631.9%
Leader: Fable 5.1 · 67.2%
knowledge
AA-Omniscience Non-hallucination
526 results- #251max of 52623.3%
- #267none of 52621.1%
- #307xhigh of 52617.6%
- #323low of 52615.7%
- #324high of 52615.6%
- #328medium of 52615.3%
Leader: MiniCPM5-1B (Non-reasoning) · 99.1%
knowledge
Harvey LAB-AA v1.1
27 results- #11max of 273.3%
Leader: Muse Wonder Full (Max) · 13.1%
- #38max of 9444.7%
Leader: Opus 5 · 63.6%
- #40max of 9683.7%
Leader: Opus 5.5 · 91.4%
- #55xhigh of 26579.7%
- #56max of 26579.7%
- #76medium of 26577.9%
- #78high of 26577.5%
- #90low of 26576.0%
- #189none of 26562.1%
Leader: Opus 5.5 · 87.7%
knowledge
Public Benefits Bench v1.1
50 results- #38max of 5057.6%
Leader: Opus 5 · 76.9%
- #21max of 52383.3%
- #80xhigh of 52380.0%
- #97high of 52379.3%
- #112medium of 52378.3%
- #170low of 52374.0%
- #344none of 52339.7%
Leader: Kimi K3 · 88.7%
- #43max of 9416.1%
Leader: Sonnet 5.5 · 75.0%
math
FrontierMath Tiers 1–3 (v2)
115 results- #16max of 11578.9%
Leader: GPT-6.1 Sol · 93.7%
- #21max of 5264.0%
Leader: Sonnet 5.5 · 100.0%
- #61max of 1161241
Leader: Fable 5 · 1308
- #10max of 851.2%
Leader: Opus 5.5 · 29.1%
- #88max of 21586.7%
- #118xhigh of 21573.0%
- #123high of 21570.3%
- #137medium of 21561.0%
- #168low of 21537.7%
- #197none of 2158.8%
Leader: Fable 5 · 98.5%
- #82max of 21659.3%
- #103xhigh of 21641.9%
- #115high of 21631.4%
- #127medium of 21618.1%
- #158low of 2164.6%
- #216none of 2160.0%
Leader: GPT-6 Astra · 95.0%
reasoning
Blueprint-Bench 2
28 results- #16max of 2831.2%
Leader: Gemini 4 · 54.4%
- #62max of 52919.4%
- #76xhigh of 52917.4%
- #88high of 52915.4%
- #113medium of 52910.6%
- #185low of 5292.6%
- #223none of 5291.1%
Leader: Muse Wonder Full (Max) · 32.9%
reasoning
Furniture Assembly
31 results- #11max of 3144.2%
Leader: Opus 5.5 · 83.3%
reasoning
Humanity's Sixth Sense
25 results- #25max of 2521.0%
Leader: GPT-6 Astra · 53.6%
reasoning
MysteryMechanism
27 results- #23max of 2719.4%
Leader: GPT-6 Astra · 53.2%
reasoning
Roboflow Visual Reasoning
102 results- #38high of 10271.1%
- #52low of 10264.2%
Leader: GPT-6 Astra · 91.2%
- #28max of 8048.1%
Leader: Opus 4.7 · 56.1%
reliability
Tool-call error rate
171 results- #41max of 1710.4%
Leader: GPT-5.4 Pro · 0.0%
- #169max of 18196.3%
Leader: GPT-5 Pro · 100.0%
- #8max of 1956.7%
Leader: Grok 4.7 · 68.3%
- #5max of 4576.3%
Leader: Wonder · 80.4%
- #21max of 322.7%
Leader: GPT-6 Astra · 56.9%
speed
Latency (Time to First Token)
351 results- #59none of 3510.9s
- #186low of 3512.5s
- #278high of 35120s
- #312xhigh of 35146s
- #328max of 351101s
Leader: Gemini 2.5 Flash-Lite · 0.3s
speed
Output Speed (tokens/s)
351 results- #104max of 351134.6
- #105high of 351133.6
- #109xhigh of 351128.3
- #113low of 351126.0
- #134none of 351112.3
Leader: Celeris-1 · 1522.5
- #134max of 1772.4s
Leader: Qwen 2.5 7B · 0.2s
- #73max of 17673.0
Leader: Muse Spark 1.1 · 204.3