# DAYJOB

> DAYJOB is Surge AI's test of long professional assignments: 80 finance tasks and 50 healthcare tasks that an agent must finish to an expert checklist. This is the combined 130-task board on Kaggle.

Canonical page: https://benchlm.ai/benchmarks/dayjob

- Category: [Agentic](/agentic)
- Last updated: Latest evaluation November 4, 2026

Evaluation results published by [Surge AI](https://www.kaggle.com/benchmarks/surge-ai/dayjob). undefined

Each task is a short workplace request backed by a multi-file workspace. An agent (OpenHands SDK harness on Kaggle) must produce finished deliverables, which a judge model grades against an expert-written checklist of binary criteria. A trial passes only when every criterion passes, and each model gets five attempts per task; the score is the share of passing trials. The combined suite has 130 tasks across 11 subfields and 8,036 rubric criteria. The healthcare and finance tracks each have their own page.

Scores are low by design: models satisfy most individual criteria but rarely all of them. Results depend on the Kaggle agent harness and the judge model, and Surge AI keeps a separate leaderboard of its own. Kaggle reports close agreement between the two, not identity.

These are published source configurations. Compare the disclosed settings before treating a score difference as a model difference.

- [Kaggle leaderboard and Apache 2.0 declaration](https://www.kaggle.com/benchmarks/surge-ai/dayjob)
- [Surge AI DAYJOB announcement](https://surgehq.ai/blog/dayjob)
- [Published leaderboard data](https://www.kaggle.com/api/v1/benchmarks/surge-ai/dayjob/leaderboard)

## About DAYJOB

- Year: 2026
- Tasks: 130 long-horizon tasks: 80 finance and 50 healthcare
- Format: Agent deliverables graded all-or-nothing against expert rubrics
- Difficulty: Long-horizon professional work

Each task is a short workplace request backed by a multi-file workspace. An agent (OpenHands SDK harness on Kaggle) must produce finished deliverables, which a judge model grades against an expert-written checklist of binary criteria. A trial passes only when every criterion passes, and each model gets five attempts per task; the score is the share of passing trials. The combined suite has 130 tasks across 11 subfields and 8,036 rubric criteria. The healthcare and finance tracks each have their own page. These results are display-only and do not enter overall or category rankings.

DAYJOB is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (19 source rows)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Claude Opus 5.5](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-03 | — | 24.5% |
| 2 | [Claude Sonnet 5.5](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-03 | — | 21.7% |
| 3 | [Gemini 4 Carbon](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | High Reasoning; Evaluated 2026-11-03 | — | 19.7% |
| 4 | [GPT-6 Astra](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-04 | — | 17.4% |
| 5 | [Gemini 4 Argon](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | High Reasoning; Evaluated 2026-11-03 | — | 15.7% |
| 6 | [GPT-6.1 Sol](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-03 | — | 15.5% |
| 7 | [Claude Opus 5](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-03 | — | 11.7% |
| 8 | [GPT-6 Sol](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-03 | — | 6.8% |
| 9 | [GPT-5.6 Sol](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-03 | — | 4.2% |
| 10 | [GPT-5.6 Terra](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | XHigh Reasoning; Evaluated 2026-11-03 | — | 2.8% |
| 11 | [Gemini 3.8 Flash](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | High Reasoning; Evaluated 2026-11-03 | — | 2.6% |
| 12 | [Claude Opus 4.8](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-03 | — | 2.0% |
| 13 | [Claude Sonnet 5](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max; Evaluated 2026-11-03 | — | 1.7% |
| 14 | [GPT-5.5](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Xhigh Reasoning; Evaluated 2026-11-03 | — | 1.4% |
| 15 | [GPT-6 Luna](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-03 | — | 1.2% |
| 16 | [GPT-5.6 Luna](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Max Reasoning; Evaluated 2026-11-03 | — | 1.1% |
| 17 | [Gemini 3 Flash Preview](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | Gemini 3 Flash High Reasoning; Evaluated 2026-11-03 | — | 0.0% |
| 18 | [Gemini 3.1 Pro Preview](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | High Reasoning; Evaluated 2026-11-03 | — | 0.0% |
| 19 | [Gemini 3.5 Flash-Lite](https://www.kaggle.com/benchmarks/surge-ai/dayjob) | High Reasoning; Evaluated 2026-11-03 | — | 0.0% |

## FAQ

### What does DAYJOB measure?

DAYJOB is Surge AI's test of long professional assignments: 80 finance tasks and 50 healthcare tasks that an agent must finish to an expert checklist. This is the combined 130-task board on Kaggle.

### Which source row has the highest reported DAYJOB value?

Claude Opus 5.5 has the highest reported DAYJOB value at 24.5%. These are published source configurations. Compare the disclosed settings before treating a score difference as a model difference.

### How many models are evaluated on DAYJOB?

The Latest evaluation November 4, 2026 contains 19 AI models.

### Does DAYJOB affect BenchLM's overall score?

Not directly. DAYJOB is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.