Benchmarks · agentic
DAYJOB: Tax
Uses the published attempt pass rate at the 95% rubric threshold across 50 professional tax assignments. Score: score. Collected by Jake Boggs' benchmark dashboard.
16 models · 16 results · data as of 2026-11-05 · surgehq.ai ↗
Leaderboard
Sonnet 5.5 max
All results
16 rows| # | Model | Effort | DAYJOB: Tax |
|---|---|---|---|
| 1 | Sonnet 5.5 | max | 10.8% |
| 2 | Fable 5.1 | max | 9.6% |
| 3 | Opus 5.5 | max | 9.2% |
| 4 | Muse Wonder (Max) | max | 6.0% |
| 5 | Grok 4.7 | xhigh | 5.0% |
| 6 | GPT-6 Astra | max | 4.8% |
| 7 | GPT-6.1 Sol | max | 4.0% |
| 8 | Muse Spark 1.3 | max | 3.6% |
| 9 | Gemini 4 Argon | high | 3.2% |
| 10 | Hy4 preview | high | 1.6% |
| 11 | DeepSeek V4.1 Flash 1017 | max | 1.2% |
| 12 | Gemini 3.8 Flash | high | 0.8% |
| 13 | GLM 5.3 | max | 0.4% |
| 14 | Kimi K3 | max | 0.0% |
| 15 | Mistral Large 4 | high | 0.0% |
| 16 | Qwen 3.8 Max | max | 0.0% |