An ultra long-horizon benchmark for measuring models on real-world Private Equity and Investment Banking workflows that would take a human professional between 20 and 30 hours to complete.
Claude Opus 5.5· max
54.8% ± 9.1
Julius Research
- Claude Opus 5.5 (max)54.8% ± 9.1
- Claude Sonnet 5.5 (max)54.6% ± 9.0
- GPT-6 Astra (xhigh)54.2% ± 8.1
- GPT-6.1 Sol (xhigh)52.9% ± 7.8
- Claude Fable 5.1 (max)45.6% ± 5.3
- GPT-6.1 Terra (xhigh)43.5% ± 6.8
- Gemini 4 Argon (high)41.3% ± 7.0
- Muse Spark 1.3 (max)37.0% ± 4.6
- Kimi K3.1 Pro (max)34.6% ± 4.4
- Grok 4.7 (xhigh)34.1% ± 4.4
- Qwen 3.8 Max (xhigh)30.3% ± 4.2
- Inkling 2 Small (max)29.8% ± 3.7
- GLM 5.3 (max)27.7% ± 3.8
- DeepSeek V4.1 Flash (max)27.3% ± 3.9
- GPT-6 Luna (xhigh)24.6% ± 4.5
- Mistral Large 4 (high)21.5% ± 4.4
- DeepSeek V4 Pro (max)20.5% ± 4.1
- Nemotron 3 Ultra (high)14.5% ± 3.3
- Inkling (max)13.1% ± 3.9
- Inkling Small (max)10.9% ± 3.8
0%10%20%30%40%50%60%70%
Mean rubric score (%)
Source: DeepFinance (Julius Research)·Updated Nov 5, 2026·juliusresearch.ai
Results: mean rubric score (%) by model configuration with its 95% confidence interval, best first
| Rank | Model | Lab | Mean rubric score with 95% confidence interval |
|---|
| 1 | Claude Opus 5.5 (max) | Anthropic | 54.8% mean rubric score, 95% CI 46.2 to 64.4 |
| 2 | Claude Sonnet 5.5 (max) | Anthropic | 54.6% mean rubric score, 95% CI 46.0 to 63.9 |
| 3 | GPT-6 Astra (xhigh) | OpenAI | 54.2% mean rubric score, 95% CI 46.4 to 62.6 |
| 4 | GPT-6.1 Sol (xhigh) | OpenAI | 52.9% mean rubric score, 95% CI 45.4 to 61.0 |
| 5 | Claude Fable 5.1 (max) | Anthropic | 45.6% mean rubric score, 95% CI 40.7 to 51.3 |
| 6 | GPT-6.1 Terra (xhigh) | OpenAI | 43.5% mean rubric score, 95% CI 37.1 to 50.6 |
| 7 | Gemini 4 Argon (high) | Google | 41.3% mean rubric score, 95% CI 34.7 to 48.8 |
| 8 | Muse Spark 1.3 (max) | Meta | 37.0% mean rubric score, 95% CI 32.6 to 41.9 |
| 9 | Kimi K3.1 Pro (max) | Moonshot AI | 34.6% mean rubric score, 95% CI 30.4 to 39.2 |
| 10 | Grok 4.7 (xhigh) | SpaceXAI | 34.1% mean rubric score, 95% CI 29.9 to 38.7 |
| 11 | Qwen 3.8 Max (xhigh) | Alibaba | 30.3% mean rubric score, 95% CI 26.3 to 34.6 |
| 12 | Inkling 2 Small (max) | Thinking Machines | 29.8% mean rubric score, 95% CI 26.1 to 33.5 |
| 13 | GLM 5.3 (max) | Z.ai | 27.7% mean rubric score, 95% CI 23.9 to 31.5 |
| 14 | DeepSeek V4.1 Flash (max) | DeepSeek | 27.3% mean rubric score, 95% CI 23.6 to 31.3 |
| 15 | GPT-6 Luna (xhigh) | OpenAI | 24.6% mean rubric score, 95% CI 20.1 to 29.2 |
| 16 | Mistral Large 4 (high) | Mistral | 21.5% mean rubric score, 95% CI 17.2 to 25.9 |
| 17 | DeepSeek V4 Pro (max) | DeepSeek | 20.5% mean rubric score, 95% CI 16.5 to 24.7 |
| 18 | Nemotron 3 Ultra (high) | NVIDIA | 14.5% mean rubric score, 95% CI 11.3 to 17.8 |
| 19 | Inkling (max) | Thinking Machines | 13.1% mean rubric score, 95% CI 9.3 to 17.2 |
| 20 | Inkling Small (max) | Thinking Machines | 10.9% mean rubric score, 95% CI 7.3 to 14.8 |
Mean rubric score by model configuration on DeepFinance (21 tasks, 10 attempts per task). Rubric score is the percentage of rubric criteria satisfied. Error bars denote 95% bootstrap confidence intervals (100,000 resamples over tasks and attempts).
Score vs. cost
Mean rubric score against mean API cost per task. The dashed line marks the Pareto frontier: configurations not dominated on both score and cost.
Sortable DeepFinance results by model configuration: mean rubric score (higher is better), API cost per task and tokens per task (lower is better)
| Rank | Model | | | |
|---|
| 1 | Claude Opus 5.5 max | 54.8 ± 9.1, 95% CI 46.2–64.4 | $8.63 | 10.5M |
| 2 | Claude Sonnet 5.5 max | 54.6 ± 9.0, 95% CI 46.0–63.9 | $6.94 | 22.6M |
| 3 | GPT-6 Astra xhigh | 54.2 ± 8.1, 95% CI 46.4–62.6 | $4.55 | 1.72M |
| 4 | GPT-6.1 Sol xhigh | 52.9 ± 7.8, 95% CI 45.4–61.0 | $0.81 | 1.90M |
| 5 | Claude Fable 5.1 max | 45.6 ± 5.3, 95% CI 40.7–51.3 | $12.02 | 3.15M |
| 6 | GPT-6.1 Terra xhigh | 43.5 ± 6.8, 95% CI 37.1–50.6 | $0.38 | 2.15M |
| 7 | Gemini 4 Argon high | 41.3 ± 7.0, 95% CI 34.7–48.8 | $2.85 | 6.69M |
| 8 | Muse Spark 1.3 max | 37.0 ± 4.6, 95% CI 32.6–41.9 | $4.87 | 10.9M |
| 9 | Kimi K3.1 Pro max | 34.6 ± 4.4, 95% CI 30.4–39.2 | $2.67 | 6.08M |
| 10 | Grok 4.7 xhigh | 34.1 ± 4.4, 95% CI 29.9–38.7 | $4.71 | 5.90M |
| 11 | Qwen 3.8 Max xhigh | 30.3 ± 4.2, 95% CI 26.3–34.6 | $4.99 | 12.6M |
| 12 | Inkling 2 Small max | 29.8 ± 3.7, 95% CI 26.1–33.5 | $0.50 | 6.38M |
| 13 | GLM 5.3 max | 27.7 ± 3.8, 95% CI 23.9–31.5 | $1.11 | 5.67M |
| 14 | DeepSeek V4.1 Flash max | 27.3 ± 3.9, 95% CI 23.6–31.3 | $0.13 | 3.93M |
| 15 | GPT-6 Luna xhigh | 24.6 ± 4.5, 95% CI 20.1–29.2 | $0.09 | 2.80M |
| 16 | Mistral Large 4 high | 21.5 ± 4.4, 95% CI 17.2–25.9 | $3.69 | 5.04M |
| 17 | DeepSeek V4 Pro max | 20.5 ± 4.1, 95% CI 16.5–24.7 | $1.50 | 2.64M |
| 18 | Nemotron 3 Ultra high | 14.5 ± 3.3, 95% CI 11.3–17.8 | $0.64 | 2.83M |
| 19 | Inkling max | 13.1 ± 3.9, 95% CI 9.3–17.2 | $0.64 | 2.40M |
| 20 | Inkling Small max | 10.9 ± 3.8, 95% CI 7.3–14.8 | $0.15 | 1.09M |
Read the post: DeepFinance: Can AI diligence a $100MM deal?