VECTOR WIREAI INTELLIGENCE
NVDA$1,847+3.2%MSFT$512+1.1%GOOGL$199-0.4%META$728+2.7%AMD$184-1.2%TSM$212+0.6%PLTR$98+4.1%AI IDX4,821+1.9%
PKT
SEEDRefresh Models Deals Regulatory Sources

TauBench V3 - Banking

100 models tracked
#ModelVendorBest scoreRunsLast seen
1Qwen3.8 Max PreviewAlibaba51.312026-08-06
2Grok 4.6SpaceXAI50.742026-08-22
3GLM-5.3Z.ai50.312026-08-19
4Qwen3.8 2.4T A95BAlibaba49.112026-08-14
5Qwen3.8 27BAlibaba4842026-08-22
6Kimi K3Moonshot4632026-07-31
7Opus 5Anthropic44.752026-07-25
8GPT-5.6 SolOpenAI44.362026-07-10
9Grok4.5SpaceXAI42.112026-07-09
10GPT-5.6 TerraOpenAI40.262026-07-10
11GPT-5.4OpenAI39.612026-07-03
12DeepSeek-V4-FlashDeepSeek39.442026-08-01
13GPT-5.5OpenAI3952026-07-03
14Claude Fable 5Anthropic38.112026-07-03
15Claude Sonnet 5Anthropic37.322026-07-03
16Gemini 3.7 FlashGoogle35.532026-08-14
17Motif-3Motif Technologies35.312026-08-13
18Muse Spark 1.2Meta34.812026-08-06
19Claude Opus 4.7Anthropic34.612026-07-03
20GLM-5.2 Full Open SourceZ.ai34.622026-07-11
21Claude Sonnet 4.6Anthropic34.412026-07-03
22Claude Opus 4.8Anthropic34.212026-07-03
23Gemini 3.5 FlashGoogle32.212026-07-03
24Muse Spark 1.1Meta31.812026-07-11
25GPT-5.6 LunaOpenAI31.162026-07-10
26DeepSeek-V4-ProDeepSeek30.122026-07-21
27Gemini 3.6 FlashGoogle29.912026-07-22
28InklingThinking Machines29.112026-07-16
29Motif-3-BetaMotif Technologies28.912026-07-21
30JT-4.1 Flash 236B A21BChina Mobile2812026-07-10
31GPT-5.4 nanoOpenAI27.412026-07-03
32Ling-3.0-flashInclusionAI27.212026-08-06
33GPT-5.4 miniOpenAI25.612026-07-03
34Claude Sonnet 4.5Anthropic24.512026-07-03
35Muse GlimmerMeta23.512026-08-11
36Solar Pro4Upstage23.312026-08-13
37Kimi K2.6Moonshot23.322026-07-07
38Hy3-previewTencent22.912026-07-14
39GPT-5OpenAI22.112026-07-05
40G9v3-39A5BAI9Stars22.112026-08-04
41Solar Open2 250BUpstage21.612026-08-13
42Gemini 3.1 ProGoogle21.422026-07-07
43DeepSeek-V3.1-TerminusDeepSeek2112026-07-09
44Gemini 3 Flash PreviewGoogle20.812026-07-03
45Ling 3.0 TinyInclusionAI20.812026-08-08
46Qwen3.6 PlusAlibaba20.822026-07-07
47Kimi K2.7 CodeMoonshot20.222026-07-07
48Muse SparkMeta19.622026-07-07
49Inkling-SmallThinking Machines18.812026-07-31
50DeepSeek-V3.2DeepSeek18.812026-07-09
51Qwen3.7 Plus PreviewAlibaba17.922026-07-07
52Gemini 3.5 Flash LiteGoogle17.512026-07-22
53Nex-N2-ProNex AGI17.522026-07-07
54Claude Sonnet 4Anthropic16.712026-07-03
55Qwen3.6 27BAlibaba16.722026-07-03
56A.X-K2SK Telecom16.112026-08-13
57GPT-5.1OpenAI15.912026-07-03
58GPT-5 miniOpenAI15.512026-07-03
59MiniMax M3MiniMax15.322026-07-07
60Qwen3.5 122B A10BAlibaba15.322026-07-09
61Gemma 4 31BGoogle14.822026-07-11
62K-EXAONELG AI14.212026-07-25
63Kimi K2.5Moonshot14.212026-07-14
64Nemotron 3 Ultra 550B A55BNVIDIA14.212026-07-03
65GLM-5.1Z.ai13.612026-07-03
66GLM-4.6Z.ai13.412026-07-03
67Qwen3.5 397B A17BAlibaba13.412026-07-03
68Grok Build 0.1SpaceXAI13.422026-07-07
69LongCat-2.0Meituan13.212026-07-20
70GPT OSS 120BOpenAI12.822026-07-14
71Agnes 2.5 Pro AlphaSapiens AI12.412026-07-25
72Grok 4.3SpaceXAI12.422026-07-03
73GLM-4.7Z.ai12.212026-07-03
74Step 3.7 FlashStepFun1222026-07-07
75Gemma 4 26B A4BGoogle1212026-07-03
76Qwen3.7 MaxAlibaba11.812026-07-03
77GPT-5.5 InstantOpenAI11.822026-07-07
78K-EXAONE 2.0 0803LG AI11.512026-08-13
79GPT-5.2OpenAI11.112026-07-03
80EXAONE 4.5 33BLG AI11.112026-07-25
81Devstral Small 2Mistral10.722026-07-07
82Devstral 2Mistral10.522026-07-07
83Nemotron 3 Super 120B A12BNVIDIA10.312026-07-03
84Nemotron Cascade 2 30B A3BNVIDIA10.112026-07-23
85MiniMax M2.7MiniMax9.922026-07-07
86Gemini 2.5 ProGoogle9.722026-07-07
87Gemini 3.1 Flash Lite PreviewGoogle9.722026-07-07
88Mercury 2Inception9.512026-07-21
89Qwen3.6 35B A3BAlibaba9.322026-07-14
90Nova 2.0 Pro PreviewAmazon9.132026-07-03
91Claude Haiku 4.5Anthropic9.112026-07-03
92Nemotron 3.5 LightningNVIDIA8.912026-08-12
93MiMo V2.5 ProXiaomi8.722026-07-07
94Gemma 4 12BGoogle8.712026-07-25
95Solar Pro 3Upstage8.722026-07-07
96Mistral Medium 3.1 (Non-Reasoning)Mistral8.212026-07-03
97Mistral Small 3.1Mistral7.422026-07-07
98LFM2.5-2.6BLiquid AI7.212026-08-22
99gpt-oss-20bOpenAI712026-07-03
100NVIDIA Nemotron 3 Nano 4BNVIDIA722026-07-08

Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.