DAYJOB, finance track (DAYJOB (Finance))
We show this table for reference; we do not rank on it.
The finance track of DAYJOB, a benchmark of end-to-end professional finance work that a model must complete as a full job task rather than a single question.
Benchmark score on DAYJOB (Finance) — November 13, 2026
We compile the DAYJOB (Finance) rows from provider self-reports. Inkling 2 Small leads the table at 14.3%. We do not use these results to rank models overall.
1 modelAgenticCurrentDisplay onlyUpdated November 13, 2026
Benchmark score table (1 model)
The one result on this page was reported by the lab that built the model.
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | Inkling 2 SmallThinking Machines LabReported by Thinking Machines Lab | 14.3% | Reported by Thinking Machines Lab | Not reported | Closed |
- Lab-reported
- Benchmark-run
- Independent
DAYJOB (Finance): reported scores versus Kaggle’s run
The highest reported DAYJOB (Finance) score in the table above is 14.3 (Inkling 2 Small). The best row in Kaggle’s own run is 25.3 (Claude Sonnet 5.5). The two cover different models: no model in the table above has a Kaggle run we can match to it, so the numbers below compare the boards, not one model under two setups.
Kaggle runs every model on the board with the benchmark authors’ code and one set of settings. A reported score uses the prompt, sampling, and reasoning settings its publisher chose, which are often not disclosed. How Kaggle Benchmarks reruns tests.
| Source | Rows | Top score | Top row | Median |
|---|---|---|---|---|
| Reported (table above) | 1 | 14.3 | Inkling 2 Small | – |
| Kaggle’s run | 19 | 25.3 | Claude Sonnet 5.5 | 3.5 |
Reported is the score in the table above. Every Kaggle row is in the Kaggle table below. Source: kaggle.com/benchmarks/surge-ai/dayjob-finance.
DAYJOB Finance (Kaggle board)
Scores are shown as published percentages. Reasoning settings, secondary metrics and evaluation dates stay on each row. Missing results are not zero. How Kaggle Benchmarks reruns tests.
| Source rank | Source model or configuration | All-pass rate (%) | Source setup | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 5.5 | 25.3% | Max Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 2 | Claude Opus 5.5 | 23.8% | Max Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 3 | Gemini 4 Carbon | 22.3% | High Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 4 | GPT-6 Astra | 21.3% | Max Reasoning; Evaluated 2026-11-04Original result | Not reported | Unknown |
| 5 | GPT-6.1 Sol | 18.5% | Max Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 6 | Gemini 4 Argon | 18.3% | High Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 7 | Claude Opus 5 | 13.3% | Max Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 8 | GPT-6 Sol | 9.3% | Max Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 9 | GPT-5.6 Sol | 6.0% | Max Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 10 | Gemini 3.8 Flash | 3.5% | High Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 11 | GPT-5.6 Terra | 3.3% | XHigh Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 12 | Claude Opus 4.8 | 2.8% | Max Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 13 | Claude Sonnet 5 | 2.8% | Max; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 14 | GPT-5.5 | 2.3% | Xhigh Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 15 | GPT-6 Luna | 1.3% | Max Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 16 | GPT-5.6 Luna | 1.0% | Max Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 17 | Gemini 3 Flash Preview | 0.0% | Gemini 3 Flash High Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 18 | Gemini 3.1 Pro Preview | 0.0% | High Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
| 19 | Gemini 3.5 Flash-Lite | 0.0% | High Reasoning; Evaluated 2026-11-03Original result | Not reported | Unknown |
How to read this leaderboard
Operator receipt: 1 sourced row is currently displayable on this page; the leading published row is Inkling 2 Small at 14.3%.
Honest limit: Low absolute scores; provider-run in the launch post.
About DAYJOB (Finance)
Year
2026
Tasks
Professional finance job tasks
Format
Task success (%)
Difficulty
Long-horizon finance work
Thinking Machines reports DAYJOB (Finance) in the Inkling 2 Small launch (November 5, 2026): Inkling 2 Small 14.3%, DeepSeek-V4.1-Flash-1017 12.5%, GLM 5.3 11.0%, GLM-5.3-Flash 7.3%, GPT-6 Luna 6.8%, Kimi K3 5.5%, Mistral Large 4 4.8% and Gemini 3.8 Flash 4.5%.
Freshness and provenance
Version
DAYJOB (Finance) 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does DAYJOB (Finance) measure?
The finance track of DAYJOB, a benchmark of end-to-end professional finance work that a model must complete as a full job task rather than a single question.
Which model scores highest on DAYJOB (Finance)?
Inkling 2 Small by Thinking Machines Lab currently leads with a score of 14.3% on DAYJOB (Finance).
How many models are evaluated on DAYJOB (Finance)?
1 AI models have published results on DAYJOB (Finance) in the BenchLM catalog.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.