# Harvey LAB — Mercor evaluation (Harvey LAB (Mercor))

> Producing legal work across practice areas and contract tasks. This table shows Mercor-run configurations for reference and is excluded from model rankings.

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

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

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

We captured Mercor's Harvey LAB results on October 15, 2026. The table preserves 40 published configurations, their source identifiers, reasoning settings, and disclosed pass@1 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 lists Harvey LAB twice on the index. We use the single linked result page once and retain its current scores, rather than combining the duplicate index cards.

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 Gemini 3 Flash Preview. Its run-count label is k = 3. The linked result page takes precedence where its setup differs, as noted above.

- [Mercor Harvey LAB results](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/)
- [Mercor APEX benchmark index](https://www.mercor.com/apex/)
- [Mercor methodology](https://www.mercor.com/apex/methodology/)
- [Upstream evaluation code](https://github.com/harveyai/harvey-labs)

## About Harvey LAB (Mercor)

- Tasks: 1,749 source-reported tasks
- Format: Pass@1
- Difficulty: Source-specific evaluation
- Paper: [Harvey LAB evaluation results](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/)

The selected headline is Pass@1. 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.

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

## Leaderboard (40 configurations)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Grok 4.7](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · xhigh reasoning · Published margin ±4.5 · 630 reported samples | xAI | 19.50% |
| 2 | [Muse Spark 1.2](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · xhigh reasoning · Published margin ±4.6 · 639 reported samples | Meta | 19.40% |
| 3 | [Opus 5](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±1.5 · 5,236 reported samples | Anthropic | 17.80% |
| 4 | [Fable 5.1](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±4.2 · 639 reported samples | Anthropic | 17.40% |
| 5 | [Muse Spark 1.3](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±4.1 · 639 reported samples | Meta | 15.00% |
| 6 | [Kimi K3](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±1.4 · 5,216 reported samples | Kimi | 14.90% |
| 7 | [GLM-5.3](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±3.8 · 639 reported samples | Zhipu | 14.60% |
| 8 | [Fable 5](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±1.4 · 5,212 reported samples | Anthropic | 14.40% |
| 9 | [Qwen3.8-Max](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · xhigh reasoning · Published margin ±1.3 · 5,204 reported samples | Alibaba | 11.90% |
| 10 | [DeepSeek-V4-Pro-0813](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±1.2 · 5,209 reported samples | DeepSeek | 11.40% |
| 11 | [Gemini 3.8 Flash](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±3.5 · 639 reported samples | Google | 11.40% |
| 12 | [GLM-5.3-Flash](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±3.4 · 638 reported samples | Zhipu | 11.10% |
| 13 | [Opus 5.5](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±3.4 · 622 reported samples | Anthropic | 10.80% |
| 14 | [Grok 4.5](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±1.2 · 5,209 reported samples | xAI | 10.70% |
| 15 | [Sonnet 5.5](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±3.6 · 545 reported samples | Anthropic | 10.00% |
| 16 | [Opus 4.8](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±1.1 · 5,227 reported samples | Anthropic | 9.40% |
| 17 | [Gemini 3.7 Flash](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±1.1 · 5,247 reported samples | Google | 9.30% |
| 18 | [GPT-6 Astra](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±3.3 · 638 reported samples | OpenAI | 9.20% |
| 19 | [GPT-6.1 Sol](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±3.4 · 616 reported samples | OpenAI | 8.80% |
| 20 | [Opus 4.7](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±1.1 · 5,236 reported samples | Anthropic | 8.70% |
| 21 | [Sonnet 5](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±1.1 · 5,188 reported samples | Anthropic | 8.60% |
| 22 | [GPT-6 Astra](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · xhigh reasoning · Published margin ±2.8 · 635 reported samples | OpenAI | 6.10% |
| 23 | [MiniMax-M3](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±0.8 · 5,209 reported samples | MiniMax | 5.30% |
| 24 | [Haiku 5.5](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±2.1 · 639 reported samples | Anthropic | 4.90% |
| 25 | [DeepSeek-V4-Pro](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±0.7 · 5,247 reported samples | DeepSeek | 4.50% |
| 26 | [Nemotron 3 Ultra](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±2.3 · 639 reported samples | NVIDIA | 4.50% |
| 27 | [GPT-5.5](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · xhigh reasoning · Published margin ±0.7 · 5,235 reported samples | OpenAI | 4.10% |
| 28 | [Gemini 3.6 Flash](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±0.8 · 5,242 reported samples | Google | 4.10% |
| 29 | [GPT-6 Sol](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±2.2 · 639 reported samples | OpenAI | 4.10% |
| 30 | [GPT-5.6 Luna](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±1.8 · 639 reported samples | OpenAI | 3.90% |
| 31 | [GPT-6 Luna](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±2 · 611 reported samples | OpenAI | 3.10% |
| 32 | [Kimi K2.7 Code](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±0.6 · 5,226 reported samples | Kimi | 2.70% |
| 33 | [GPT-5.6 Sol (Pro)](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±0.6 · 5,134 reported samples | OpenAI | 2.70% |
| 34 | [GLM-5.1](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · Published margin ±1.6 · 639 reported samples | Zhipu | 2.70% |
| 35 | [GPT-5.6 Terra](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · max reasoning · Published margin ±1.4 · 639 reported samples | OpenAI | 2.30% |
| 36 | [MiniMax-M2.7](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±0.4 · 5,244 reported samples | MiniMax | 1.90% |
| 37 | [Gemini 3.5 Flash Lite](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±1 · 639 reported samples | Google | 1.30% |
| 38 | [GPT-5.4](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · xhigh reasoning · Published margin ±0.4 · 5,242 reported samples | OpenAI | 1.20% |
| 39 | [Gemini 3.5 Flash](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±0.3 · 5,345 reported samples | Google | 0.70% |
| 40 | [Gemini 3.1 Pro](https://www.mercor.com/apex/partner-benchmarks/harvey-lab-leaderboard/) | Harvey LAB Agent · high reasoning · Published margin ±0.1 · 5,247 reported samples | Google | 0.10% |

## FAQ

### What does Harvey LAB (Mercor) measure?

Producing legal work across practice areas and contract tasks. This table shows Mercor-run configurations for reference and is excluded from model rankings.

### Which model leads the published Harvey LAB (Mercor) snapshot?

Grok 4.7 currently leads the published Harvey LAB (Mercor) snapshot with a score of 19.50%.

### How many models are evaluated on Harvey LAB (Mercor)?

The October 15, 2026 capture contains 40 source configurations.

### Does Harvey LAB (Mercor) affect BenchLM's overall score?

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