# GeneBench-Pro — Mercor evaluation (GeneBench-Pro (Mercor))

> Taking messy genomics datasets through quality control, modeling and diagnostics to a verifiable answer. This table shows Mercor-run configurations for reference and is excluded from model rankings.

Canonical page: https://benchlm.ai/benchmarks/mercor-genebench-pro

- Category: [Agentic](/agentic)
- Last updated: November 12, 2026 capture

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

We captured Mercor's GeneBench-Pro results on November 12, 2026. The table preserves 36 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.

Mercor scores the 10-problem public set of OpenAI's GeneBench-Pro, with ten runs per model on each problem, so margins are wide. The scores labs report on their own runs of the benchmark are listed separately and are not combined with this table.

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 binary: the final answer is checked against a recoverable target within calibrated tolerances and a judge of Programatic answer checking. Its run-count label is k = 10. The linked result page takes precedence where its setup differs, as noted above.

- [Mercor GeneBench-Pro results](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/)
- [Mercor APEX benchmark index](https://www.mercor.com/apex/)
- [Mercor methodology](https://www.mercor.com/apex/methodology/)
- [Source paper](https://www.biorxiv.org/content/10.64898/2026.06.29.735386v2.full.pdf)
- [Upstream dataset](https://huggingface.co/datasets/openai/genebench-pro-public-package)

## About GeneBench-Pro (Mercor)

- Tasks: 10 public problems
- Format: Mean Score
- Difficulty: Source-specific evaluation
- Paper: [GeneBench-Pro evaluation results](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/)

Mercor runs the 10-problem public set of OpenAI's GeneBench-Pro, ten times per model on each problem. The headline is the mean score. Scores the labs report on the full benchmark stay on the GeneBench-Pro page.

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

## Leaderboard (36 configurations)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Sonnet 5.5](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±28.5 · 99 reported samples | Anthropic | 39.00% |
| 2 | [GPT-6 Astra](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±23.5 · 100 reported samples | OpenAI | 36.00% |
| 3 | [GPT-6.1 Sol](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±23.5 · 100 reported samples | OpenAI | 33.00% |
| 4 | [Opus 5.5](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±30 · 100 reported samples | Anthropic | 30.00% |
| 5 | [Muse Wonder 1.0](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±30 · 20 reported samples | Meta | 30.00% |
| 6 | [Muse Spark 1.3](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±23.5 · 100 reported samples | Meta | 26.00% |
| 7 | [Fable 5.1](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±26 · 100 reported samples | Anthropic | 26.00% |
| 8 | [GPT-6 Sol](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±24 · 100 reported samples | OpenAI | 26.00% |
| 9 | [GPT-5.6 Sol](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±19 · 100 reported samples | OpenAI | 23.00% |
| 10 | [Opus 5](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±21 · 100 reported samples | Anthropic | 22.00% |
| 11 | [GLM-5.3](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±17.5 · 100 reported samples | Zhipu | 15.00% |
| 12 | [Grok 4.6](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · xhigh reasoning · Published margin ±18 · 100 reported samples | xAI | 15.00% |
| 13 | [Fable 5](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±17.5 · 100 reported samples | Anthropic | 14.00% |
| 14 | [Gemini 3.7 Flash](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · high reasoning · Published margin ±17 · 100 reported samples | Google | 14.00% |
| 15 | [GPT-6 Luna](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±16 · 100 reported samples | OpenAI | 13.00% |
| 16 | [DeepSeek-V4.1-Flash](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±9.5 · 100 reported samples | DeepSeek | 11.00% |
| 17 | [Grok 4.5](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · high reasoning · Published margin ±14.5 · 100 reported samples | xAI | 11.00% |
| 18 | [Opus 4.7](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±15.5 · 100 reported samples | Anthropic | 11.00% |
| 19 | [Gemini 3.8 Flash](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · high reasoning · Published margin ±15.5 · 100 reported samples | Google | 11.00% |
| 20 | [Gemini 3.6 Flash](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · high reasoning · Published margin ±14 · 100 reported samples | Google | 10.00% |
| 21 | [Grok 4.7](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · xhigh reasoning · Published margin ±15 · 100 reported samples | xAI | 10.00% |
| 22 | [Sonnet 5](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±11 · 100 reported samples | Anthropic | 10.00% |
| 23 | [Muse Spark Open](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · xhigh reasoning · Published margin ±15 · 20 reported samples | Meta | 10.00% |
| 24 | [DeepSeek-V4-Flash](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±10.5 · 100 reported samples | DeepSeek | 9.00% |
| 25 | [Gemini 3.5 Flash](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · high reasoning · Published margin ±12.5 · 100 reported samples | Google | 9.00% |
| 26 | [Kimi K3](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±6.5 · 100 reported samples | Kimi | 6.00% |
| 27 | [GPT-5.6 Terra](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±5.5 · 100 reported samples | OpenAI | 5.00% |
| 28 | [MiniMax-M3](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · high reasoning · Published margin ±5 · 100 reported samples | MiniMax | 4.00% |
| 29 | [GPT-5.5](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · xhigh reasoning · Published margin ±6 · 100 reported samples | OpenAI | 4.00% |
| 30 | [Beam](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±4 · 100 reported samples | Reflection | 4.00% |
| 31 | [GPT-5.6 Luna](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±3.5 · 100 reported samples | OpenAI | 3.00% |
| 32 | [Qwen3.8-Max](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · xhigh reasoning · Published margin ±3 · 100 reported samples | Alibaba | 2.00% |
| 33 | [Inkling](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · high reasoning · Published margin ±1.5 · 100 reported samples | Thinking Machines | 1.00% |
| 34 | [Nemotron 3 Ultra](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · high reasoning · Published margin ±1.5 · 100 reported samples | NVIDIA | 1.00% |
| 35 | [DeepSeek-V4-Pro-0813](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · max reasoning · Published margin ±1.5 · 100 reported samples | DeepSeek | 1.00% |
| 36 | [Gemini 3.5 Flash Lite](https://www.mercor.com/apex/oss-benchmarks/genebench-leaderboard/) | Terminus 2 · high reasoning · Published margin ±0 · 100 reported samples | Google | 0.00% |

## FAQ

### What does GeneBench-Pro (Mercor) measure?

Taking messy genomics datasets through quality control, modeling and diagnostics to a verifiable answer. This table shows Mercor-run configurations for reference and is excluded from model rankings.

### Which model leads the published GeneBench-Pro (Mercor) snapshot?

Sonnet 5.5 currently leads the published GeneBench-Pro (Mercor) snapshot with a score of 39.00%.

### How many models are evaluated on GeneBench-Pro (Mercor)?

The November 12, 2026 capture contains 36 source configurations.

### Does GeneBench-Pro (Mercor) affect BenchLM's overall score?

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