# APEX-Accounting — Mercor evaluation (APEX-Accounting (Mercor))

> Month-end accounting tasks that require work across files and applications. This table shows Mercor-run configurations for reference and is excluded from model rankings.

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

- Category: [Agentic](/agentic)
- Last updated: October 15, 2026 capture

Evaluation results published by [Mercor](https://www.mercor.com/apex/apex-accounting-leaderboard/). Published evaluation aggregates only; task contents and dataset license grants are not included.

We captured Mercor's APEX-Accounting results on October 15, 2026. The table preserves 34 published configurations, their source identifiers, reasoning settings, and disclosed mean score values. These are results from Mercor's evaluation setup; we did not rerun them.

This dated source table is display only and does not enter overall or category rankings. Source configurations retain their external identity, including Pro descriptors and different effort settings.

Mean score is the percentage of rubric criteria passed; Pass@1 requires all criteria for a task to pass. The dedicated accounting page describes 160 tasks, the Loop harness, a three-attempt mean-criteria headline, and 95% task-level bootstrap intervals from 10,000 resamples. Its eight-attempt reliability description and the general methodology's four-run single-shot description differ. Three legacy scalar records omit a harness and sample count; their labels retain that missing information. This table remains separate from Ramp Accounting Bench.

The row labels retain published error margins and sample counts when supplied. Missing margins and counts remain unreported; a sample count is not used to reconstruct a task denominator. Except where the dedicated source defines an interval, the confidence level is unreported. Equal displayed scores and overlapping margins do not establish statistical ties. Capture time does not establish individual evaluation dates.

The general methodology lists expert-authored rubrics, scored criterion by criterion and a judge of LLM judge over expert rubrics. Its run-count label is k = 4. The linked result page takes precedence where its setup differs, as noted above.

- [Mercor APEX-Accounting results](https://www.mercor.com/apex/apex-accounting-leaderboard/)
- [Mercor APEX benchmark index](https://www.mercor.com/apex/)
- [Mercor methodology](https://www.mercor.com/apex/methodology/)
- [Source paper](https://arxiv.org/abs/2607.27189)
- [Mercor evaluation notes](https://www.mercor.com/blog/introducing-the-ai-productivity-index-for-accounting/)

## About APEX-Accounting (Mercor)

- Tasks: 160 held-out accounting tasks
- Format: Mean Score
- Difficulty: Source-specific evaluation
- Paper: [APEX-Accounting evaluation results](https://www.mercor.com/apex/apex-accounting-leaderboard/)

The selected headline is Mean Score. Source model identifiers, effort settings, published margins, and reported sample counts remain attached to the source configuration. Public-set results and Mercor task extensions occupy separate tables.

APEX-Accounting (Mercor) is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (34 configurations)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Opus 5.5](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±4 · Pass@1 14.8% ±4.9 · 637 reported samples | Anthropic | 61.80% |
| 2 | [Fable 5.1](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.7 · Pass@1 11.7% ±4.3 · 639 reported samples | Anthropic | 61.00% |
| 3 | [Opus 5.5](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · medium reasoning · Published margin ±3.7 · Pass@1 12.7% ±4.6 · 640 reported samples | Anthropic | 59.90% |
| 4 | [GPT-6 Astra](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.8 · Pass@1 11.9% ±4.5 · 640 reported samples | OpenAI | 57.90% |
| 5 | [Fable 5](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.6 · Pass@1 9.8% ±4 · 480 reported samples | Anthropic | 56.40% |
| 6 | [Opus 5](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.9 · Pass@1 10.5% ±3.9 · 640 reported samples | Anthropic | 54.50% |
| 7 | [Muse Spark 1.3](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.7 · Pass@1 10.8% ±3.9 · 630 reported samples | Meta | 53.50% |
| 8 | [GLM-5.3](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.7 · Pass@1 9.2% ±3.3 · 639 reported samples | Zhipu | 53.10% |
| 9 | [Muse Spark 1.1](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · xhigh reasoning · Published margin ±3.6 · Pass@1 7.7% ±3.1 · 480 reported samples | Meta | 52.60% |
| 10 | [Gemini 3.8 Flash](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · high reasoning · Published margin ±3.9 · Pass@1 8.7% ±3.7 · 640 reported samples | Google | 51.70% |
| 11 | [GPT-5.6 Sol](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Harness not reported for this record · max reasoning · Published margin ±3.9 · Pass@1 8.4% ±3.5 · Sample count not reported | OpenAI | 51.50% |
| 12 | [GPT-5.6 Sol (Pro)](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.9 · Pass@1 9.6% ±4.1 · 480 reported samples | OpenAI | 51.50% |
| 13 | [Muse Spark 1.2](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · xhigh reasoning · Published margin ±4 · Pass@1 8.9% ±3.6 · 599 reported samples | Meta | 51.40% |
| 14 | [Kimi K3](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.6 · Pass@1 6.2% ±2.7 · 635 reported samples | Kimi | 49.90% |
| 15 | [GPT-5.5](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · xhigh reasoning · Published margin ±3.6 · Pass@1 6.2% ±3 · 633 reported samples | OpenAI | 48.20% |
| 16 | [DeepSeek-V4.1-Flash](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.6 · Pass@1 7.8% ±3.4 · 640 reported samples | DeepSeek | 48.10% |
| 17 | [Opus 4.8](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.6 · Pass@1 4.2% ±2.5 · 477 reported samples | Anthropic | 48.00% |
| 18 | [GPT-5.4](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · xhigh reasoning · Published margin ±3.8 · Pass@1 5.3% ±2.7 · 637 reported samples | OpenAI | 46.80% |
| 19 | [Grok 4.7](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · xhigh reasoning · Published margin ±3.1 · Pass@1 4.2% ±1.8 · 638 reported samples | xAI | 46.70% |
| 20 | [Grok 4.5](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · high reasoning · Published margin ±3.5 · Pass@1 4.5% ±2.3 · 631 reported samples | xAI | 43.70% |
| 21 | [Opus 4.6](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.5 · Pass@1 4.4% ±2.7 · 616 reported samples | Anthropic | 43.30% |
| 22 | [GLM-5.2](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Harness not reported for this record · Published margin ±3.7 · Pass@1 4% ±1.8 · Sample count not reported | Zhipu | 42.70% |
| 23 | [GPT-5.6 Terra](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.9 · Pass@1 6% ±2.6 · 634 reported samples | OpenAI | 40.30% |
| 24 | [MiniMax-M3](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · high reasoning · Published margin ±3.3 · Pass@1 2.2% ±1.2 · 623 reported samples | MiniMax | 39.10% |
| 25 | [GPT-5.6 Luna](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±3.9 · Pass@1 4.4% ±2.6 · 629 reported samples | OpenAI | 38.00% |
| 26 | [Kimi K2.7 Code](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Harness not reported for this record · auto reasoning · Published margin ±3.3 · Pass@1 2.9% ±1.7 · Sample count not reported | Kimi | 37.00% |
| 27 | [Gemini 3.1 Pro](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · high reasoning · Published margin ±3 · Pass@1 3.1% ±2.1 · 622 reported samples | Google | 34.00% |
| 28 | [Qwen3.5](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · Published margin ±2.9 · Pass@1 0% ±0 · 471 reported samples | Alibaba | 24.40% |
| 29 | [Inkling](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · high reasoning · Published margin ±2.8 · Pass@1 0.8% ±0.9 · 638 reported samples | Thinking Machines | 24.40% |
| 30 | [Nemotron 3 Ultra](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · high reasoning · Published margin ±3 · Pass@1 0.3% ±0.4 · 602 reported samples | NVIDIA | 22.90% |
| 31 | [DeepSeek-V3.2](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · Published margin ±2.8 · Pass@1 1.1% ±0.8 · 636 reported samples | DeepSeek | 20.70% |
| 32 | [GLM-5.1](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · Published margin ±4.3 · Pass@1 1.2% ±1.3 · 266 reported samples | Zhipu | 18.60% |
| 33 | [GLM-5.3-Flash](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · max reasoning · Published margin ±2.8 · Pass@1 1.7% ±1.4 · 640 reported samples | Zhipu | 10.80% |
| 34 | [GPT-OSS-120B](https://www.mercor.com/apex/apex-accounting-leaderboard/) | Loop · high reasoning · Published margin ±1 · Pass@1 0% ±0 · 636 reported samples | OpenAI | 2.10% |

## FAQ

### What does APEX-Accounting (Mercor) measure?

Month-end accounting tasks that require work across files and applications. This table shows Mercor-run configurations for reference and is excluded from model rankings.

### Which model leads the published APEX-Accounting (Mercor) snapshot?

Opus 5.5 currently leads the published APEX-Accounting (Mercor) snapshot with a score of 61.80%.

### How many models are evaluated on APEX-Accounting (Mercor)?

The October 15, 2026 capture contains 34 source configurations.

### Does APEX-Accounting (Mercor) affect BenchLM's overall score?

Not directly. APEX-Accounting (Mercor) is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.