Tau Banking (Artificial Analysis)

Agentic

Source ↗ Higher is better

Rank	Model	Provider	Score
1	Qwen3.8 Max	Alibaba	51.3%
2	Grok 4.6 (high)	SpaceXAI	50.7%
3	Muse Spark 1.3 (max)	Meta	50.5%
4	GLM-5.3 (max)	Z.ai	50.3%
5	Qwen: Qwen3.8 2.4T A95B	Alibaba	49.1%
6	Qwen3.8 27B (xhigh)	Alibaba	48.0%
7	Qwen: Qwen3.8 Max (0902)	Alibaba	47.8%
8	Qwen3.8 27B (medium)	Alibaba	47.4%
9	Muse Spark 1.3 (xhigh)	Meta	47.2%
10	Z.ai: GLM 5.3 Flash	Z.ai	47.2%
11	Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)	Anthropic	47.2%
12	Kimi K3 (max)	Moonshot AI	46.0%
13	Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)	Anthropic	45.8%
14	Gemini 3.8 Flash (medium)	Google	45.8%
15	Qwen3.8-Flash-Next	Alibaba	45.4%
16	Gemini 3.8 Flash (high)	Google	44.9%
17	Claude Opus 5 (Adaptive Reasoning, High Effort)	Anthropic	44.7%
18	Grok 4.6 (medium)	SpaceXAI	44.3%
19	GPT-5.6 Sol (max)	OpenAI	44.3%
20	Grok 4.6 (xhigh)	SpaceXAI	43.3%
21	Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)	Anthropic	43.3%
22	GPT-6 Astra (xhigh)	OpenAI	43.1%
23	Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)	Anthropic	43.1%
24	Claude Opus 5 (Adaptive Reasoning, Max Effort)	Anthropic	42.1%
25	Grok 4.5 (high)	SpaceXAI	42.1%
26	Kimi K3 (low)	Moonshot AI	41.6%
27	GPT-6 Astra (max)	OpenAI	41.4%
28	DeepSeek V4 Flash Vision (Reasoning, Max Effort)	DeepSeek	41.0%
29	Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)	Anthropic	41.0%
30	GPT-5.6 Terra (max)	OpenAI	40.2%
31	GPT-6 Astra (high)	OpenAI	40.0%
32	DeepSeek V4 Pro 0813 (Reasoning, Max Effort)	DeepSeek	39.6%
33	GPT-5.4 (xhigh)	OpenAI	39.6%
34	DeepSeek V4 Flash 0731 (Reasoning, Max Effort)	DeepSeek	39.4%
35	GPT-5.5 (xhigh)	OpenAI	39.0%
36	Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)	Anthropic	39.0%
37	Claude Opus 5 (Adaptive Reasoning, Medium Effort)	Anthropic	38.6%
38	inclusionAI: Ling 3.0 Flash FinHistorical	inclusionAI	38.6%
39	GPT-5.6 Sol (xhigh)	OpenAI	38.1%
40	Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)	Anthropic	38.1%
41	Grok 4.6 (low)	SpaceXAI	38.1%
42	Claude Sonnet 5 (Adaptive Reasoning, Max Effort)	Anthropic	37.3%
43	GPT-5.6 Sol (high)	OpenAI	36.7%
44	GPT-5.5 (high)	OpenAI	36.7%
45	GPT-5.6 Sol (medium)	OpenAI	36.5%
46	Agnes 2.5 Pro Beta	Sapiens AI	35.7%
47	GPT-6 Astra (medium)	OpenAI	35.5%
48	Gemini 3.7 Flash (medium)	Google	35.5%
49	Motif 3	Motif Technologies	35.3%
50	Muse Spark 1.2 (xhigh)	Meta	34.8%

Access this data

JSON GET /api/v1

[
  {
    "rank": 1,
    "model": "Qwen3.8 Max",
    "provider": "Alibaba",
    "score": "51.340206185567"
  },
  {
    "rank": 2,
    "model": "Grok 4.6 (high)",
    "provider": "SpaceXAI",
    "score": "50.7216494845361"
  },
  {
    "rank": 3,
    "model": "Muse Spark 1.3 (max)",
    "provider": "Meta",
    "score": "50.5154639175258"
  }
]

Curl

$ curl -s https://rightrank.ai/api/v1/rankings/tau-banking-artificial-analysis.json | jq '.[:3]'

CLI brew install rightrank

$ rightrank top tau-banking-artificial-analysis --limit 5
$ rightrank compare claude-opus-4.6 gpt-5.4-pro --category tau-banking-artificial-analysis
$ rightrank value tau-banking-artificial-analysis  # best performance per dollar

Claude Code Skill Skill

# In Claude Code, use the RightRank skill:
/rightrank best model for tau-banking-artificial-analysis
/rightrank compare top 5 by value
/rightrank cheapest model scoring above 80 on tau-banking-artificial-analysis