MCP tool orchestration

Results are only compared within a matching group, so a model tested one way is never ranked against a model tested another way.

Coordinate MCP tools to complete a task

Ranking groups use separate scales. Compare configurations only inside the same group; EvalRank never flattens unlike systems into one leaderboard.

ベンチマーク横断のコンセンサス

このトピックで採用されているベンチマーク全体でモデルを順位付けしたものです。意思決定のときと同じ重み付けを用いています。各スコアは、モデルの結果が最高結果に占める割合です。

RankModelShare of bestBenchmarks
1GPT-6 Sol100.0%1
2GPT-5.6 Sol97.2%2
3Claude Sonnet 5.594.9%1
4Claude Haiku 5.574.0%1
5Claude Sonnet 570.2%1
6Claude Sonnet 4.664.2%2
7claude-haiku-4-534.8%2

エージェントシステム · エージェンティック

Exploratory - not a ranked winner yet
rg-mcp-tool-orchestration-agent-system-agentic-agent-system-v1

エクスプローラーのエビデンス

AutomationBench

Sources differ; shown separately, never averaged。に確認しました。まで有効です。

RankConfigurationCapabilityUncertaintyFamilies
1Gemini 4 Carbon (High) automationbench:Gemini 4 Carbon (High)config_e84fd0f5e59e3b8b964a542fe8b56ecab0908e9f39f2d70113f5958cfa087e47100%不明1
2Gemini 4 Carbon (Medium) automationbench:Gemini 4 Carbon (Medium)config_a2c8a23afe15a50ed53869a323929661fea1b94adf65bc273c11b4020ed5df2a98%不明1
3Gemini 4 Argon (High) automationbench:Gemini 4 Argon (High)config_5d2ea9047d28198e8f9d59299190aa9c7aa87d2bd564c2115aa0707b1f5d587695%不明1
4Gemini 4 Argon (Medium) automationbench:Gemini 4 Argon (Medium)config_75cc09df9b6c3d3d3e18e6383c21bc35c08dd06ca09b77f568d02e92dfd2182d93%不明1
5Claude Sonnet 5.5 (default fallbacks, Max) automationbench:Claude Sonnet 5.5 (default fallbacks, Max)config_24df20a13eb0b425ca163f80c22071b19068661c34ded9d3b9e570596851664083%不明1
6Claude Opus 5.5 (default fallbacks, Max) automationbench:Claude Opus 5.5 (default fallbacks, Max)config_7d573508be54b283cb5b9da147bca47fbbe0b1999ac7fb96b2188a220b9e089d79%不明1
7GPT-6 Astra (Max) automationbench:GPT-6 Astra (Max)config_a111367f67b8b5bd55e3f1ba436f00f5ad0576721d4f7d1e6c4143e11c591bb577%不明1
8GPT-6 Astra (XHigh) automationbench:GPT-6 Astra (XHigh)config_1f8e8f6eec27fbfec971a587be15a5f5ff62ee7697d7d822c2958b926c2ff34e72%不明1
9GPT-6 Astra (High) automationbench:GPT-6 Astra (High)config_790ab0f3fbf9884524df53588326d812de8d04023ac15a466f32581d4e65192c69%不明1
10Claude Sonnet 5.5 (default fallbacks, XHigh) automationbench:Claude Sonnet 5.5 (default fallbacks, XHigh)config_38411a32bd75fcf1e407e91922f08b0fce52b9b484b9245cc728060bdae2438d68%不明1
11Claude Opus 5.5 (default fallbacks, XHigh) automationbench:Claude Opus 5.5 (default fallbacks, XHigh)config_d156e9fba8f9e0c88442fbdd4c718e9be1cf24a73d7dae582bf97215d95028f566%不明1
12GPT-6 Astra (Medium) automationbench:GPT-6 Astra (Medium)config_926a2bc702bb0bb28519705811f091560eb26aa4a9bb8be86ff75cac1d16a48963%不明1
13GPT-6 Sol (XHigh) automationbench:GPT-6 Sol (XHigh)config_bcd7cba856ee883231d52bd67acbc695263f9d9b6d30f84eaeace38956512f4562%不明1
14Claude Opus 5.5 (default fallbacks, High) automationbench:Claude Opus 5.5 (default fallbacks, High)config_0ebb36576624ae257dc2e45c695763430113c68d0d679619c79e9a4354157e8e61%不明1
15GPT-6 Sol (Max) automationbench:GPT-6 Sol (Max)config_0ae6fc70172dbbe05e35a00fc5dd331f291406d91e47eea51030184315c38fb259%不明1
16Gemini 4 Carbon (Low) automationbench:Gemini 4 Carbon (Low)config_0022478ede674058059e374503c9b6e6bcf75d2a928e33fd243a27ffd862464758%不明1
17GPT-6 Sol (High) automationbench:GPT-6 Sol (High)config_d91199fe2cf3c7cd2c612bd66f5f8e17b5e0d2c0f8693d73d8ca14f15c3ee3f158%不明1
18Gemini 3.7 Flash (High) automationbench:Gemini 3.7 Flash (High)config_c0cb9d036f2945081943c28d51317f9d58b56a8414efe3190d392ddf789049a156%不明1
19GPT-6 Astra (Low) automationbench:GPT-6 Astra (Low)config_bbddcff9f0e33fbae7e6300bccdbfcd194fb626b52162cb7076b546464af9d6f56%不明1
20Gemini 3.8 Flash (Medium) automationbench:Gemini 3.8 Flash (Medium)config_095b858200422ae2a40f4de8999825fc1169b7634be76069bcf1630b1c43b24c55%不明1
21Gemini 3.8 Flash (High) automationbench:Gemini 3.8 Flash (High)config_6fb0bae145783ee3b353b0ca54772a85f9a0286483bbeb389733f8693880832255%不明1
22Claude Opus 5.5 (default fallbacks, Medium) automationbench:Claude Opus 5.5 (default fallbacks, Medium)config_80d1f934366a7bb43b583184ba8e185bc373d803668d1a190a1eedf251603a8555%不明1
23Claude Sonnet 5.5 (default fallbacks, High) automationbench:Claude Sonnet 5.5 (default fallbacks, High)config_d63c3d216dfc81713b0aad7a4df86eecf61dac1a3b70f7d6533ee2c968b8877555%不明1
24GPT-5.6 Sol (Max) automationbench:GPT-5.6 Sol (Max)config_940a9985646aa2382eb4b6250e00321dbd4848b72f450cb4f78d19d7dbb641ec53%不明1
25Claude Opus 5 (Max) automationbench:Claude Opus 5 (Max)config_6c30ed4edc61c38064a7dfb3e0367ca8c4cdb51598fe51220192c028246cc3f750%不明1
26GPT-6 Sol (Medium) automationbench:GPT-6 Sol (Medium)config_8991690c485ea7dcfea93958c8eba8f59139a26a7986edbf7874bce3350c00e250%不明1
27GPT-5.6 Sol (XHigh) automationbench:GPT-5.6 Sol (XHigh)config_6914105bab6b84c5f7857a2daf10cece7fcb4d8dddd505c553f87f034e308fc749%不明1
28Claude Opus 5 (XHigh) automationbench:Claude Opus 5 (XHigh)config_379a60c5001bd6e347dcb9bba5e3b43039bb97f4873a41c6a9cef13654870ccf47%不明1
29GPT-6 Astra (None) automationbench:GPT-6 Astra (None)config_e0b5ab861bbb87d5302dfdcbf53d1cdbd54665212c0848975a0be745fc54e14e47%不明1
30GPT-5.6 Sol (High) automationbench:GPT-5.6 Sol (High)config_48d7667275573b7f84817bffb5ec6ec29dacc0cab92c3c5221335df4c4bb778e46%不明1
31Claude Opus 5.5 (default fallbacks, Low) automationbench:Claude Opus 5.5 (default fallbacks, Low)config_800d2388f620e5a5f20eef6c3d8634c88201853f5fad8a2fad124433dd44f85d45%不明1
32Claude Opus 5 (Medium) automationbench:Claude Opus 5 (Medium)config_38af236b2377faa1da2903ec87e4e8f14bcf8b96ff6dd9f2fffacf69b30e9c0144%不明1
33GPT-5.6 Terra (Max) automationbench:GPT-5.6 Terra (Max)config_66ab823f13f439e05e8d8001792d1c4c28d3ec56bfe2dd9d3bf63dd1407d2cfd44%不明1
34Claude Sonnet 5.5 (default fallbacks, Medium) automationbench:Claude Sonnet 5.5 (default fallbacks, Medium)config_3560eaf095efa2d9a30365236cec46651041d20016cbf77cf4616c66f6b134ba43%不明1
35Kimi K3 automationbench:Kimi K3config_7fe334a4ba41ed28eb6687619f824748ebd7d4cbe9e99cbbc796d00233f7ff2b42%不明1
36Claude Fable 5.1 (Max) automationbench:Claude Fable 5.1 (Max)config_5a05961823ba6f70a3e2e768a7f45bfd863c22abfd6cd358f58566ed229d2c8342%不明1
37Claude Fable 5.1 (XHigh) automationbench:Claude Fable 5.1 (XHigh)config_7f4f479be11ddadd0afd2f3cfbb3488555fed2308346fe2e0749498ba7c6cd7d40%不明1
38GPT-6 Sol (Low) automationbench:GPT-6 Sol (Low)config_1bd5e38efb3ba147b60840c08f238ca2f71746d64ea30531aa0693e53c4a938e39%不明1
39Muse Spark 1.3 (Max) automationbench:Muse Spark 1.3 (Max)config_843ac9cf1d323628644076897150b6dbb112c5dc07c08c5a953d78dba9e6df0d38%不明1
40GPT-6 Luna (Max) automationbench:GPT-6 Luna (Max)config_b76316b06957619144608fa6606d170063ba394dd5f95b0f5749ae897bca667338%不明1
41Claude Opus 5 (High) automationbench:Claude Opus 5 (High)config_ee7ea30c3c59725eadd7a39119f3a97212ec84a86ab3f654f8d64c73e8e2786c38%不明1
42Claude Opus 5 (Low) automationbench:Claude Opus 5 (Low)config_b4fd15298892317616ae4d34fd5dd4ab6048d96d82232bdaee2783bc4e4fba1a38%不明1
43GPT-5.6 Sol (Medium) automationbench:GPT-5.6 Sol (Medium)config_c72ba89a8964da6640cd2110b783cdfedb335585b95cd3c6b91d0deab5c8f2e836%不明1
44Claude Sonnet 5.5 (default fallbacks, Between tools) automationbench:Claude Sonnet 5.5 (default fallbacks, Between tools)config_dd8b98785baf8f3f1260e2caa1f113add821b2615aee2ff6c405b4865a9372ea34%不明1
45Claude Sonnet 5.5 (default fallbacks, Low) automationbench:Claude Sonnet 5.5 (default fallbacks, Low)config_035c383289bf40bf0aa358453f36f0328f095913f12bc4ad3cd1569b3357183534%不明1
46DeepSeek V4 Flash (Max) automationbench:DeepSeek V4 Flash (Max)config_8adec4fa3f4445669babf6791fd0c8b1d13faaf33fe2e1b5ab77a8578a4b8ae134%不明1
47Claude Fable 5.0 (Max) automationbench:Claude Fable 5.0 (Max)config_842a0156d3bc02b07e5557c6f76d8817f36c71821facb520fbdc89c69ff7a0c832%不明1
48GPT-5.6 Luna (Max) automationbench:GPT-5.6 Luna (Max)config_9f407c81e3e732a87ee9966b14be9fed65ccbfd04e208da88a597efc81348ffa32%不明1
49Gemini 3.6 Flash (High) automationbench:Gemini 3.6 Flash (High)config_edbc6630272a0a2a0d726c090798ce08f4aefc75f0768687904fe6d52bae8ef132%不明1
50Claude Opus 4.8 (XHigh) automationbench:Claude Opus 4.8 (XHigh)config_57a659c9e83d821868dab27ab0c159c57b5c97fb718ac66d09d1c5fd5f6f6f8731%不明1
51GPT-5.5 (XHigh) automationbench:GPT-5.5 (XHigh)config_a44921f0a694e40f05bbaf53e494a883740be650b703bdeb2e6d88247204cc9131%不明1
52Claude Opus 4.8 (Max) automationbench:Claude Opus 4.8 (Max)config_6fc288bc8171bcb11a4d115a1236362d6f2c1cace393b9455376b1a7446a6bd031%不明1
53Gemini 3.6 Flash (Medium) automationbench:Gemini 3.6 Flash (Medium)config_9883ff967e78a4bc23fc63fa7515b224b645585ffa07f0adb4793b3e321de2a031%不明1
54Claude Fable 5.0 (High) automationbench:Claude Fable 5.0 (High)config_1e09d3f93509113695dc72682b6d4bd1a220ac6f8973c216ab0268c5ac6e85d330%不明1
55Claude Opus 4.8 (High) automationbench:Claude Opus 4.8 (High)config_7af054dd8b90178d41fa5ba45ac6b85a7bd9843dcaecf1792dae961b753dc77930%不明1
56Gemini 3.6 Flash (Low) automationbench:Gemini 3.6 Flash (Low)config_96db8eff7c7487fa8b3edcc5a5b6749f8a26631031b5987c4fab2f292049e97129%不明1
57Claude Fable 5.0 (XHigh) automationbench:Claude Fable 5.0 (XHigh)config_9e1bb3935c837e86f05fbcdb05fbce0a02636a323176f9ee7dd106488e279d2928%不明1
58GPT-5.6 Terra (XHigh) automationbench:GPT-5.6 Terra (XHigh)config_0938f8c812774ae0f3301e7e73f7da3d97672ad7ad16a0e5328e641a5a91c82c27%不明1
59GPT-6 Luna (High) automationbench:GPT-6 Luna (High)config_bca94f1f9298e81fe8ba14b154d2a09ce9a1508b6ee4861dd5df64ad82bfd7e427%不明1
60Claude Opus 4.8 (Medium) automationbench:Claude Opus 4.8 (Medium)config_9bc17cc37c8ee20959203a8f4947c5b3af8bc2609a8396087cdbaa58c01d45af26%不明1
61Claude Fable 5.0 (Medium) automationbench:Claude Fable 5.0 (Medium)config_016ff96a0e4331eec7742f8a9d4fd190c38091cc7b4e5bc176cd0f3d68a1b0ee25%不明1
62Claude Opus 4.8 (Low) automationbench:Claude Opus 4.8 (Low)config_1933e3ab343725efb7037ee8c4dc8cb0825f9decf507e06c6690479aeb34940f25%不明1
63Claude Opus 4.7 (Max) automationbench:Claude Opus 4.7 (Max)config_59e570755a4cf5d7fbb07f086d044c1bbf95cfa2dbf629a504a5ca5ef00d9ddf25%不明1
64Gemini 3.5 Flash (High) automationbench:Gemini 3.5 Flash (High)config_755449f4b0a662664c5dbafb0c1d0dbeb840e0293ef0ca007b51d69549ca956125%不明1
65GPT-5.6 Luna (XHigh) automationbench:GPT-5.6 Luna (XHigh)config_b8209af48673e711712a4e514aa2a6cfa4d0b1f8fee6c10d13dc8ac9dd117e3024%不明1
66GPT-5.5 (High) automationbench:GPT-5.5 (High)config_8eff39f6fbfb5661ca1f0191233bafda3a85e8b642e4070bea355b251519831b24%不明1
67GPT-6 Luna (XHigh) automationbench:GPT-6 Luna (XHigh)config_e34d98131c28ce73764be376543465683a40ccd737a87830ffec3e5937ce96be23%不明1
68Claude Fable 5.0 (Low) automationbench:Claude Fable 5.0 (Low)config_538922a0f6a51bf97e124666493ed18ea7f0329ba5ad951d9cb396bcbc20979023%不明1
69Claude Haiku 5.5 (default fallbacks, Max) automationbench:Claude Haiku 5.5 (default fallbacks, Max)config_2d730b70968a8445ec73c8964d1d773d21bfdca141a8aecd61edc914ef38870922%不明1
70GPT-5.6 Terra (High) automationbench:GPT-5.6 Terra (High)config_683e62a9dd0be7c3ce9e8da6b7a1fcf7591ada7d4dd9e0206f7fe96677b1f9d222%不明1
71GPT-5.6 Sol (Low) automationbench:GPT-5.6 Sol (Low)config_98207ee200454b8ea84e956f71ecba3113d7f36d19749b06da86c646b055fc0122%不明1
72GPT-5.4 (XHigh) automationbench:GPT-5.4 (XHigh)config_e8064f3e3a4c1e8d61ed769cf52efb0672f7816498084d140131ffa26193195a21%不明1
73Gemini 3.5 Flash (Medium) automationbench:Gemini 3.5 Flash (Medium)config_26955bb84631e73a688c108a6155adfdc11dd856fe5c262f1d36057bb7b3542a21%不明1
74Claude Sonnet 5 (Max) automationbench:Claude Sonnet 5 (Max)config_972f27eeef575bb98d84efd9cfa4a1f3d1159d8210431dc30a14b7070ff906b920%不明1
75Claude Haiku 5.5 (default fallbacks, High) automationbench:Claude Haiku 5.5 (default fallbacks, High)config_e7d0cc2bffe97d13bcd84dc6661c180b1b596bfb9511ead4345d21bb132d36c619%不明1
76GPT-5.4 (High) automationbench:GPT-5.4 (High)config_f9a4242ec20cdc9f97ad292ac4ac3ad675cbfa489dd1392154e088019786c3b319%不明1
77Claude Sonnet 5 (High) automationbench:Claude Sonnet 5 (High)config_20282f5cf0dbd7c112d93547824df5fcdee2cde63ea002164dd4701f113cf64218%不明1
78Claude Haiku 5.5 (default fallbacks, XHigh) automationbench:Claude Haiku 5.5 (default fallbacks, XHigh)config_3554b3b0c77b31f9b29d994dc9b7dec9286af6e127eca2d7b9b21e74a9f6097818%不明1
79Claude Opus 4.7 (XHigh) automationbench:Claude Opus 4.7 (XHigh)config_4358b0b845f5bdb1fc0fe74e682a7e6f8611b28ef6e9020d7cfc79b317926dc818%不明1
80GPT-6 Luna (Medium) automationbench:GPT-6 Luna (Medium)config_aaed56f1623126e9e0f3c9f83fe8fbf57f088abe7a8ea052159749bea11aacb918%不明1
81GPT-5.6 Luna (High) automationbench:GPT-5.6 Luna (High)config_6d2cc4180e12e117508f5272e4612a8a92945c4b96779a26a8f0ae0becd12c3c17%不明1
82Claude Opus 4.7 (High) automationbench:Claude Opus 4.7 (High)config_de903b17293f654071e06d06bbd68685a6b8b1daf141a1536c8828e4506ef7df17%不明1
83GPT-6 Sol (None) automationbench:GPT-6 Sol (None)config_0b6eb43d309326b5f507fb96fd383a58ccf83eb6b8d0a617de9be5081c7d0de717%不明1
84Gemini 3.5 Flash (Low) automationbench:Gemini 3.5 Flash (Low)config_231bbc69d64c7ecdc426b5ddca802275a979a51fc9c79074919b3122df0802c917%不明1
85GPT-5.5 (Medium) automationbench:GPT-5.5 (Medium)config_656eec9a117375a7a1d64d5d93991d50d899a608624f25c56a5d2c89818e4c1617%不明1
86Gemini 3.1 Pro (preview) (High) automationbench:Gemini 3.1 Pro (preview) (High)config_cfdc83f500f5f2027e4b83edc59e0701a1c000bb9645a20874170899130e1dd716%不明1
87Claude Haiku 5.5 (default fallbacks, Medium) automationbench:Claude Haiku 5.5 (default fallbacks, Medium)config_9f8d37e89ebdeb515dc7d6ec7db89a53eedd3872a4414d8abda399b9f7fcc55716%不明1
88GPT-5.6 Terra (Medium) automationbench:GPT-5.6 Terra (Medium)config_f4ab754d979d37b2482d7f87e1438f5495ac5188a8cfeb8c4d1f4fceec5bdf3a15%不明1
89GLM 5.2 (Max) automationbench:GLM 5.2 (Max)config_0c13ca0ba185e27a9f5c82afde759268a5f5e77f5594464c0103d3957d614a9914%不明1
90Claude Sonnet 5 (XHigh) automationbench:Claude Sonnet 5 (XHigh)config_46217d1aac5ae584b99d820b707bf9a7ced9949c46d9b327d5de4da4df08aab014%不明1
91GPT-5.6 Terra (Low) automationbench:GPT-5.6 Terra (Low)config_617e8371d486c79f25c0e07ae993e3495c113b65fc930005ee07b38fb44deec614%不明1
92Gemini 3.1 Pro (preview) (Medium) automationbench:Gemini 3.1 Pro (preview) (Medium)config_c8fd43f830cbdb9b4503da89b1db1ec30caa30f3879ebae54294322105660a5b13%不明1
93Claude Opus 4.7 (Medium) automationbench:Claude Opus 4.7 (Medium)config_2800b26b583f7d5d02436aae781338e2a96e54e82429d8191e16676583b5de1a13%不明1
94GLM 5.1 (Max) automationbench:GLM 5.1 (Max)config_55964024fb11ee6532d7554782d28f013f1056dbf08ebb8c9bd447189e6cfb7312%不明1
95Claude Haiku 5.5 (default fallbacks, Low) automationbench:Claude Haiku 5.5 (default fallbacks, Low)config_34ae6300e749bf6a300ffb78bd3ee3c48e0a47d5c09df05e1daa18a8777d97bd11%不明1
96Claude Sonnet 5 (Medium) automationbench:Claude Sonnet 5 (Medium)config_69f2227cb119fd66a3ec31face79684824168194551dc155b990cebd7ed9f33d11%不明1
97Claude Opus 4.7 (Low) automationbench:Claude Opus 4.7 (Low)config_0ace5839fcad4044d3df332bae99e48417cbbf6744e139c5c78b4e135428ccf110%不明1
98Claude Sonnet 4.6 (Max) automationbench:Claude Sonnet 4.6 (Max)config_b98e2fee1d3cf0c74de948a3d0d45b5b80c9f51a16a7f5d660c823a1cb29222610%不明1
99GPT-5.4 (Medium) automationbench:GPT-5.4 (Medium)config_d1966837232621d34f8fc04ca0a4b7e80b692c23a077aa498455ead06822bd8110%不明1
100Claude Sonnet 4.6 (High) automationbench:Claude Sonnet 4.6 (High)config_917f2ef386a3a6346baf40eb79c1e64adcd06b4922ac984800b0ce54f1171f519%不明1
101Minimax M3 (Max) automationbench:Minimax M3 (Max)config_0b34173771cf7e524bf382430c1f0bddec294a39a446c29f1d2838f536d6f2c99%不明1
102Gemini 3.1 Pro (preview) (Low) automationbench:Gemini 3.1 Pro (preview) (Low)config_9c231cb0712e622189c62e9d503de44ce2e978d4bb9970b0b1cdcaa60df629f68%不明1
103Qwen 3.6+ (High) automationbench:Qwen 3.6+ (High)config_ac8c95d3767eb2450487c2de81bb2dd53fee30be7c5348cdad6aaa714dd0f03d8%不明1
104GPT-5.6 Luna (Medium) automationbench:GPT-5.6 Luna (Medium)config_7a9ec1209ea915c2f075866e5c9e7e496774b0e4098e9cecadfa82fc822470858%不明1
105Gemini 3.5 Flash Lite (Medium) automationbench:Gemini 3.5 Flash Lite (Medium)config_5965182fe9b7ac2b4b4bb68e3a68f28e12fa95580a161dff9e32b8cbbb17b45e7%不明1
106GPT-5.6 Sol automationbench:GPT-5.6 Solconfig_0fe3df83d5405cb879e7c888d3ec219c37f8abfb1b77fac891c4d1a83bec7ef27%不明1
107Qwen 3.7+ automationbench:Qwen 3.7+config_b1ab5e4e13101409eda689ce8df061d2b2c0d934660398fa80ddea334e3250a27%不明1
108Claude Sonnet 4.6 (Medium) automationbench:Claude Sonnet 4.6 (Medium)config_60e7088d0f31383097dd602e14d9c3293c966054d0a1a86a59362b9a01100b837%不明1
109Claude Sonnet 5 (Low) automationbench:Claude Sonnet 5 (Low)config_d512e3d3de66475a32922dd7c1fcea8ff173d6c57fb3126d644cafbcd373d3db7%不明1
110Gemini 3.5 Flash Lite (High) automationbench:Gemini 3.5 Flash Lite (High)config_a1f5d3b72d69f9325d23335544f0cb70a1b4ef6115866fd4f4e2ebfbe10ba1946%不明1
111Kimi K2.7 Code automationbench:Kimi K2.7 Codeconfig_ab845b1f3ab15c72ebe2f75b680321bc084498e7d4c7e7fdf618ce0d0e9df3076%不明1
112Claude Sonnet 4.6 (Low) automationbench:Claude Sonnet 4.6 (Low)config_0ccde52aa509afd4e2350cb38a101a54b7665b73a20654c5a17a66aaa7271c055%不明1
113GPT-5.5 (Low) automationbench:GPT-5.5 (Low)config_24e8d53c0136d8b5629ecff77e4455c68b17328ff9f8296451a1cbd32c3bfc0b5%不明1
114Kimi K2.6 automationbench:Kimi K2.6config_f459a46c670d91cee6f393e92df639704024fd609fd9772213160ed6b2f0382b5%不明1
115GPT-5.6 Luna (Low) automationbench:GPT-5.6 Luna (Low)config_f3f80ec5f6bcba5acda6376458367925ac1495556eb56c1f3de677a3f4b14d563%不明1
116Gemma 4 31B (Max) automationbench:Gemma 4 31B (Max)config_76f665c645a8364c51556168df78bf0eea23138b7a2b5da4677778f80647fdd43%不明1
117GPT-5.4 (Low) automationbench:GPT-5.4 (Low)config_4fb251b393d11b9386f207b0f37d67759be0dcaed5ecbcbcb0b77896ba4606e53%不明1
118GPT-6 Luna (Low) automationbench:GPT-6 Luna (Low)config_4570dbe7a5543ef8a8248ae5c61dc2ebee528e2938861a3dab8c36e9c93250fe2%不明1
119Gemini 3.5 Flash (Minimal) automationbench:Gemini 3.5 Flash (Minimal)config_b54c11ea31d7647b3d7db5177afe3a157affea280a4d973fb311e13a5ac35fb42%不明1
120GPT-5.5 automationbench:GPT-5.5config_11e954943fd9ad77cabe35d72b06200fa98db7508157154de7c3785df66058972%不明1
121GPT-5.6 Terra automationbench:GPT-5.6 Terraconfig_cf03f5b3c4e7a90d8a90130659a125a2377620b1d4b57ce89ccd6d5b184c9c6a1%不明1
122Claude Haiku 4.5 automationbench:Claude Haiku 4.5config_99e5bf5740dd30017b10fe9e3fe1e4ea8edb3f10d66879ee9bbfa545d4f1ef141%不明1
123Minimax M2.7 (High) automationbench:Minimax M2.7 (High)config_bea74616b427750ea86b97daee9f4973d8894f2958d5cb66a35740ab3e41eb0c1%不明1
124GPT-6 Luna (None) automationbench:GPT-6 Luna (None)config_415054aa14c4ebcb5967ebb9c853eb09f6f5951d2b3334f17783a5d0fa7e19cc1%不明1
125GPT-5.4 automationbench:GPT-5.4config_5471863f8d76f77203a971b328349f3a41a8c9c6b3fb7f436e38250d2bcd80a00%不明1
126GPT-5.6 Luna automationbench:GPT-5.6 Lunaconfig_bc7a49651846ff7d1b19f8a048bc5bf5f309f59aea304331361eb342ea17a0af0%不明1
127GPT-OSS 120B (High) automationbench:GPT-OSS 120B (High)config_cfa46f246cd66e02d2b0dd618d8de1f72d1e999d10abe5661068575dfbb636a40%不明1
128Gemini 3.5 Flash Lite (Low) automationbench:Gemini 3.5 Flash Lite (Low)config_f9644e1833ae570913681da98d4ba6f799add0287f54495d4696bc9704bd28580%不明1
このグループの構成を比較する

エクスプローラーのエビデンス

MCP-Atlas

Sources differ; shown separately, never averaged。に確認しました。まで有効です。

RankConfigurationCapabilityUncertaintyFamilies
1Muse Spark 1.1 mcp-atlas:Muse Spark 1.1config_865e6ef688acad27a266666361d822a7e12a1942834149471cf90eb044a5cc19100%不明1
2claude-Fable 5.1 mcp-atlas:claude-Fable 5.1config_f328e3998ed1b45103e1649d2a56eb49ab52c4778c1c7fbb601e4a09c6aeaa8498%不明1
3claude-opus-5 mcp-atlas:claude-opus-5 | xhighconfig_3b3da0e8d86c658bad97c185bb9a483596e0071070d7cf422d13e9cc9fe0b05a95%不明1
4Qwen3.8-2.4T-A95B mcp-atlas:Qwen3.8-2.4T-A95B | xhighconfig_1390b81eca2217ffd224a50bf934dcc42987b13c9d626c93327d94eb0f77af7892%不明1
5GLM 5.3 mcp-atlas:GLM 5.3config_4230f907d1dac8530ad295039e60329f300e8b5c03fbdce4de791a419dedd21492%不明1
6Gemini 3.5 Flash mcp-atlas:Gemini 3.5 Flash | highconfig_b5619ab8c5a5e56ed4fb07192538b4c60398967a98bd1e84aa003fd40b5e470191%不明1
7Claude Fable 5 mcp-atlas:Claude Fable 5config_0af42acc3be8b535fe45e8528a110115e3dd4285b0a8d6240ec15d79e7e0a66390%不明1
8kimi-k3 mcp-atlas:kimi-k3 | maxconfig_6f65c5cb596ddad3df6a940b2b8290cd820b7461521faa187b025da51bb12bbc88%不明1
9Muse Spark mcp-atlas:Muse Sparkconfig_47028847b57705edc5e8485a9c561b1f1088e683eee397c8f02e5aca140727c788%不明1
10claude-opus-4-8 mcp-atlas:claude-opus-4-8 | maxconfig_56ece83507303ffc5a54b226e22f8386a83c748a7904b50812c53e8ad254d61688%不明1
11gpt-5.6-sol mcp-atlas:gpt-5.6-solconfig_14e638ae9ebf6ea3d9836679b084df5917ab75b83fc026d187f1b5fc48b4a1ce87%不明1
12Inkling-small mcp-atlas:Inkling-smallconfig_f75f6d17e3c366b14091c9625505c7c70b115a94c65fc825da57566170ece92b81%不明1
13claude-opus-4-7 mcp-atlas:claude-opus-4-7 | maxconfig_05a06b37c0a4784f1186582e4860942edb8279ac8f2c9a12392cfa5971436ecf81%不明1
14gemini-3.1-pro-preview mcp-atlas:gemini-3.1-pro-preview | highconfig_d16487ba4140f90ab8e629b2bd15a10683a8a0b38a5b422addc585163fa8fc2b79%不明1
15glm-5p2 mcp-atlas:glm-5p2config_9699b17040d7a7e000ce022b080619085249b28ef899665d8db4775a89f737fc78%不明1
16claude-opus-4-6 mcp-atlas:claude-opus-4-6 | maxconfig_07f910f0cb8559b0faf9b23dcb787e5fdbc4c3f879f396903e7bd95686c37c8076%不明1
17Inkling mcp-atlas:Inkling | xhighconfig_40527c8c703ea129249d2e9b7e67687c3ef8098f305b5caec0730ef571e53d5d75%不明1
18glm-5p1 mcp-atlas:glm-5p1config_b052e66e22d1ee4d2bad78bd6113209e193d9488c4c8167aa1d24acfb8e5813c74%不明1
19gpt-5.5 mcp-atlas:gpt-5.5 | xhighconfig_f5077b4a3b6a10c5fb4ceb2f793cfdc2cac38a5eb5a489a8a3c9aac250172ca673%不明1
20gpt-5.4 mcp-atlas:gpt-5.4 | xhighconfig_0f3ec8178be2beb726efa9e54d9bbe8b0ebe163b046451b1c0e618665453b12963%不明1
21gemini-3-pro-preview mcp-atlas:gemini-3-pro-previewconfig_83f12c7c544421b5ca5d1de95594c62fae3c76cf7eccba162e4debc2541c8e6463%不明1
22claude-opus-4-5 mcp-atlas:claude-opus-4-5 | highconfig_3554bcad8ceadb2c86d5eef891263f8fd1aa56bae525786b724d3f03c2d2c25362%不明1
23claude-sonnet-4-6 mcp-atlas:claude-sonnet-4-6config_4b53a75c0bf72e6e78ceb7f3710a73def172cecfd1481e27fcd95a9c19efbcf961%不明1
24gpt-5.2 mcp-atlas:gpt-5.2 | xhighconfig_880a3ad28d5bd24cc745c1e6bf260f3d3d196263b0d8cdf285f6e5956e60581857%不明1
25kimi-k2p5 mcp-atlas:kimi-k2p5config_fba0ee6e849a5c93b689b9d0e3b26034b06b7a2cf9ee2222d6dd7bae9ed30c2351%不明1
26Nemotron 3 Ultra mcp-atlas:Nemotron 3 Ultra | thinkingconfig_90a7dd9d9a4e3539f3d56132efa93b8433610735bd9985371cdf6cb9d9c9087948%不明1
27gemini-3-flash-preview mcp-atlas:gemini-3-flash-previewconfig_c55ef9457cb3ed93270f0f5fd9bde56665511b7d532751c17944fee26cfff27246%不明1
28claude-sonnet-4-5 mcp-atlas:claude-sonnet-4-5 | thinkingconfig_6bff94df52f572fd58c975cffac1e3f415ea59f18493e9d86d141a7a4eb0b83940%不明1
29glm-4p7 mcp-atlas:glm-4p7config_4f4efb7aa85ec2fa61d79749a4c2219af212f5fd4df9f0b743c3ac807193597b37%不明1
30gemini-3.1-flash-lite mcp-atlas:gemini-3.1-flash-lite | highconfig_42937da441ce651eb4640549352dc34c2ff35e1c9d9aa539e3fab35e9a28ad9d35%不明1
31gpt-5.4-mini mcp-atlas:gpt-5.4-mini | xhighconfig_5131da2f10f8f911dabcc4468860eda25d94ced53ea86759194bac58c9236b1434%不明1
32gpt-5.1 mcp-atlas:gpt-5.1 | highconfig_5cc4d6801b3c13b8d9033fac9aa87f665bd0a7383b61f77e8fbb0c1de9892ed921%不明1
33o3-pro mcp-atlas:o3-proconfig_e8fdd869d71f2e1bc97d8ea15b2b753bc9b579a1aad2976d16370aca6735df0c9%不明1
34claude-haiku-4-5 mcp-atlas:claude-haiku-4-5config_0980c78f9c8c89f5a9484df46ff794b35518b66cbdb0a81dce709b25a7594c9d0%不明1
このグループの構成を比較する

エクスプローラーのエビデンス

tau3-bench Banking Knowledge

Sources differ; shown separately, never averaged。に確認しました。まで有効です。

RankConfigurationCapabilityUncertaintyFamilies
1Qwen 3.8 Max qwen3-8-max_sierra_2026-08-04config_a26a8d10f6a1f8c55b2ae39ab85459ab0c65f1c000c357819e67050c18dcc2fc55%不明1
2Claude Opus 5 claude-opus-5_sierra_2026-08-04config_47e4a643a1c879897e35acc19a52271766e0d24716490d433f47f1250db796f549%不明1
3Grok 4.5 grok-4-5_sierra_2026-08-04config_f4ada0ce1a4b8539b6b16eb70ecd5937576c5612b57153c438fab65dc2644d9648%不明1
4GPT-5.6-sol gpt-5-6-sol_sierra_2026-08-04config_a3711cbeeacfc54666ebdd477195ebcaac4642a7aaff878de8a40aa19416a56847%不明1
5GPT-5.5 gpt-5-5_sierra_2026-05-05config_ad5041576a055588650eb979cd38244d3c92d872d052d82dafb205bfa74cad3d45%不明1
6Muse Spark 1.1 muse-spark-1-1_sierra_2026-08-04config_9e9d0ff85e658a60ac616eb6536f77d57ceb67178820283b459b21fc91e3716440%不明1
7Claude Opus 4.7 claude-opus-4-7_sierra_2026-05-05config_442ebf6d5800bf504193b385c4554a54f85fd95ef7845c2c54708c68861e021640%不明1
8Claude Opus 4.8 claude-opus-4-8_sierra_2026-08-04config_53dbefe80b9536d86ae8cb24809c569e4c07be1cca283343280e6be8bb56314540%不明1
9Claude Fable 5 claude-fable-5_sierra_2026-08-04config_7bf49a074c69d34202fcdb4efe6bd1931566fb4e29b9f6e773de585366ecc67b40%不明1
10GPT-5.4 gpt-5-4_sierra_2026-03-25config_81277ce76a2443b26ca7f3db93d65526b5c66a673148fbe228d16db598379eb739%不明1
11Kimi K3 kimi-k3_sierra_2026-08-04config_0ac9cb2463f2a18ab6613660534e55deb75861bf471271f6aac39d02e4fbc4f837%不明1
12GLM-5.2 glm-5-2_sierra_2026-08-04config_ebd02a3bcc5de5fdc15841e5214662ff73054096aefd3b9db9de3c1ced3cc67937%不明1
13GPT-5.2 gpt-5-2_sierra_2026-02-26config_0a61de28f7e96ed842d0988f4ff17d52f778051137caca3b12c8809fab3a1e2432%不明1
14Distyl ButtonAgent distyl-buttonagent_distyl_2026-03-25config_91e6242dafc69cf6d5b045dfa3e30dbca4c38c586b61a1ecbb28962e03b9d7c631%不明1
15Claude Opus 4.6 claude-opus-4-6_sierra_2026-05-05config_79e8f1ddf8bcb22e452c0322f0a78c5fbcba059bfa5acc9acce62e5702eee47227%不明1
16Gemini 3 Flash gemini-3-flash_sierra_2026-03-02config_8095fa144e2c9e0944193c86724a27a3b80ddc5fb3d62d561fa298fa54c7970127%不明1
17Gemini 3.1 Pro Preview gemini-3-1-pro-preview_sierra_2026-05-05config_03d738b190a028db163266b395b4eca0e0f719f7bb43bd1a6db305d10f91fe2d26%不明1
18Claude Sonnet 4.5 claude-sonnet-4-5_sierra_2026-02-26config_12ecf4a9f0ae1e2f9815530d97599aef123564d64fddecf8fd39aad2165b1a6e25%不明1
19Inkling inkling_sierra_2026-08-04config_4a49384540eefc80eaff83393407119b8902549d0f6a45bed272b60bd4ee493725%不明1
20Claude Opus 4.5 claude-opus-4-5_sierra_2026-02-26config_311ca40d213a79344ba2b618fe7f29fcf71ff8cbaaa8cab632499a1bf6dc01b025%不明1
21Grok 4.2 grok-4-2_sierra_2026-05-05config_62cd4f1736a5056e6cf24954b6e0fe45123ac500dcbe4726080840da176bc5a818%不明1
22Gemini 3 Pro gemini-3-pro_sierra_2026-03-02config_66ed191ca632aad42f3b0c528e2912ccef29a49ac134594c4f0209ac45e77ef718%不明1
23Grok 4 fast grok-4-fast_sierra_2026-05-05config_3835bdd87e33611308716fcc185637e9d15136981937b1d92c25d7ae2977592716%不明1
24Gemini 2.5 Pro gemini-2-5-pro_sierra_2026-05-05config_5908158689798bbd956792b26fbe7a5a831a7907a7f811a7e974b7231763d00b14%不明1
25Grok 4.1 fast grok-4-1-fast_sierra_2026-05-05config_320872c79fbebdc15ccb655f4f54fd38971fc9049a5bdb2a2ce5209a2fc755ea13%不明1
26GPT-5.2 gpt-5-2-none_sierra_2026-02-26config_414bd51eab2ed44504e875d40d7cfe5b9246c80ca5bb91ca01accabf5625ea9713%不明1
27GLM-5 glm-5-think_sierra_2026-03-02config_3f1ad20f11e4674c490744ee115c4bc0c88728936353a52de2d451677b63812f10%不明1
28Qwen3.5-397B-A17B qwen3.5-397b-a17b-think_sierra_2026-03-02config_54cb0b0fed47395e14ff29324863729b3a12e1f93ad171469ba3b588e56dd17710%不明1
このグループの構成を比較する

エクスプローラーのエビデンス

Toolathlon

Sources differ; shown separately, never averaged。に確認しました。まで有効です。

RankConfigurationCapabilityUncertaintyFamilies
1GLM 5.3 Flash toolathlon-verified:GLM 5.3 Flash | Default | reasoning=maxconfig_a1aadfcb56405a0d0eb191f5135a9d9e2ccb7dee2c238801790214226a6a78ba78%不明1
2Kimi K3 toolathlon-verified:Kimi K3 | Default | reasoning=maxconfig_3c68f7e1a7f2486c9ff1b6472b4c236cc5336d62caaede33a34a94c8b18131ba77%不明1
3Claude Opus 4.8 toolathlon-verified:Claude Opus 4.8 | Default | reasoning=maxconfig_40bbb60a17931a21d62e2e9759806e9ee686d1024058b5e441fa991ef351117576%不明1
4Muse Spark 1.2 toolathlon-verified:Muse Spark 1.2 | Default | reasoning=xhighconfig_9feb5dad2c8a17db8a8a081b11fc4f030ecd8b0058bcdf81d49c2fc78dcf1f0a76%不明1
5Muse Spark 1.1 toolathlon-verified:Muse Spark 1.1 | Default | reasoning=xhighconfig_b3dd3426004e01579b0fa24d5e2d3af6a1055eea5369d7be8cb7a3bbbc76785f76%不明1
6DeepSeek V4 Pro 0813 toolathlon-verified:DeepSeek V4 Pro 0813 | Default | reasoning=maxconfig_a442fb1aa154afa78f31f9333e14fbcceeb68d81bc8ff09f12cbf0a76563a7b774%不明1
7GPT-5.5 toolathlon-verified:GPT-5.5 | Default | reasoning=xhighconfig_14a00cacba7d65e688a5c760163b8a5c43f157cd0ac3c74a8908ada4bf75af4a74%不明1
8Claude Sonnet 5 toolathlon-verified:Claude Sonnet 5 | Default | reasoning=maxconfig_1e92cd632b85b5cae4db98f0747a379d6971e1901e817b6c70ba668a1d4f258172%不明1
9DeepSeek V4 Flash 0731 toolathlon-verified:DeepSeek V4 Flash 0731 | Default | reasoning=maxconfig_088be908339ec2c5d18efb1d89f10f2a589d84061a7df91df36136544658b0fe71%不明1
10Gemini 3.5 Flash toolathlon-verified:Gemini 3.5 Flash | Default | reasoning=highconfig_7424c21364f1c26da7c03d4246703a4574f36bcf742ec051d026f1a2866292a567%不明1
11Gemini 3.1 Pro toolathlon-verified:Gemini 3.1 Pro | Default | reasoning=highconfig_6387719edef57bba35460d38ec2768197f35735c9d27dfbc49a4dacf8b8749c461%不明1
12GLM 5.2 toolathlon-verified:GLM 5.2 | Default | reasoning=maxconfig_32d1f9502ae274db028b789ff6ea67f263fbea181b3f45dd6ab9a33c4032113860%不明1
13Kimi K2.7 Code toolathlon-verified:Kimi K2.7 Code | Defaultconfig_013440bcabc69495c6ad1bfdd9fbf082239f7167ccc9dc6f3a721aacb614434158%不明1
14Kimi K2.6 toolathlon-verified:Kimi K2.6 | Defaultconfig_c6aa2be8da38947cda5c3c1ab5ce5ba57e8dd5b1a105e0734ec8b3d785cf6e9a58%不明1
15Gemini 3.5 Flash-Lite toolathlon-verified:Gemini 3.5 Flash-Lite | Default | reasoning=highconfig_425f38f920acec6ada6502aa651d29309804f062d7ed03c8eed3c54fa62c02d557%不明1
16Hy3 toolathlon-verified:Hy3 | Default | reasoning=highconfig_446911b3349e7d3b5aad2cacc1d0b27500170fc47e6e2a36c433ae9723f807b457%不明1
17DeepSeek V4 Pro toolathlon-verified:DeepSeek V4 Pro | Default | reasoning=maxconfig_7527b100f7599e56734b703f9af93a6879ba5d71dd9e6ee4c136a0814017202356%不明1
18Inkling-Small toolathlon-verified:Inkling-Small | Default | reasoning=xhighconfig_2ed2bf4206c9be5f348242ccb5526dc4a6b0195f657a6b33e34f7be422924e5d54%不明1
19DeepSeek V4 Flash toolathlon-verified:DeepSeek V4 Flash | Default | reasoning=maxconfig_f8a8eb20465f68c1be43c3ad9d4e9c2d41a80a4425f90ea4071fd5ba0355acf851%不明1
20MiMo V2.5 toolathlon-verified:MiMo V2.5 | Defaultconfig_5fbd46f4fa4376922f5b7517a6d938344057a9542a0428d8b0f62cef6a1b2f2d49%不明1
21MiniMax M2.7 toolathlon-verified:MiniMax M2.7 | Defaultconfig_60f8d3600b9bcb5b8877cbd43ef898f2662079d65b4a7db91dc9319f8507dee648%不明1
22Inkling toolathlon-verified:Inkling | Default | reasoning=xhighconfig_cb28eb8804eaba125c2e9fce39868f34be7b2e47fb2da8261a4f54b3f73d542746%不明1
23Qwen3.5 397B-A17B toolathlon-verified:Qwen3.5 397B-A17B | Defaultconfig_69d341d1f12b8f7988cf1cbd0f72515cb5fab5ddcb250c6ce72a038de5a66e0041%不明1
24Nemotron 3 Ultra toolathlon-verified:Nemotron 3 Ultra | Defaultconfig_e3ea78145d4d26cc32da136c704897302f33b6b7cb7d7beba53fcc66f7be9c5e34%不明1
25Kimi K2.5 toolathlon-verified:Kimi K2.5 | Defaultconfig_8be3ad7082404c320dce37e31ba992fae9c1f78835fc94cef5b7b4e2504672a533%不明1
このグループの構成を比較する
上位セットを定めるための独立したエビデンスが不足しています。
  • キャリブレーションが検証されていません。
  • 構成の重複が基準値を下回っています。
  • 独立したベンチマークファミリーのカバレッジが基準値を下回っています。
Back to all kinds of work