# APEX-SWE (Terminus-2) — Mercor evaluation (APEX-SWE (Terminus-2) (Mercor))

> Software integration and production-failure diagnosis with the Terminus-2 harness. This table shows Mercor-run configurations for reference and is excluded from model rankings.

Canonical page: https://benchlm.ai/benchmarks/mercorapexsweterminus2

- Category: [Coding](/coding)
- Last updated: October 15, 2026 capture

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

We captured Mercor's APEX-SWE (Terminus-2) results on October 15, 2026. The table preserves 57 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.

These are the disclosed Terminus-2 Pass@1 results across integration and observability work. Inspect is a separate harness profile. Configurations without a score for the selected harness stay absent, rather than receiving zero.

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 plus repository test suites and a judge of LLM judge + programmatic tests. Its run-count label is k = 3. The linked result page takes precedence where its setup differs, as noted above.

- [Mercor APEX-SWE (Terminus-2) results](https://www.mercor.com/apex/apex-swe-leaderboard/)
- [Mercor APEX benchmark index](https://www.mercor.com/apex/)
- [Mercor methodology](https://www.mercor.com/apex/methodology/)
- [Source paper](https://arxiv.org/pdf/2601.08806)
- [Upstream evaluation code](https://github.com/Mercor-Intelligence/apex-swe)
- [Upstream dataset](https://huggingface.co/datasets/mercor/APEX-SWE)
- [Mercor evaluation notes](https://www.mercor.com/blog/introducing-apex-swe/)

## About APEX-SWE (Terminus-2) (Mercor)

- Tasks: Integration and observability engineering tasks
- Format: Pass@1
- Difficulty: Source-specific evaluation
- Paper: [APEX-SWE (Terminus-2) evaluation results](https://www.mercor.com/apex/apex-swe-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.

APEX-SWE (Terminus-2) (Mercor) is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (57 configurations)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Opus 5.5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±5.9 · 800 reported samples | Anthropic | 67.60% |
| 2 | [Sonnet 5.5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.1 · 800 reported samples | Anthropic | 66.40% |
| 3 | [Opus 5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.4 · 800 reported samples | Anthropic | 63.70% |
| 4 | [Fable 5.1](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.3 · 776 reported samples | Anthropic | 63.60% |
| 5 | [Fable 5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.4 · 799 reported samples | Anthropic | 58.80% |
| 6 | [Grok 4.6](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · xhigh reasoning · Published margin ±6.2 · 1,600 reported samples | xAI | 57.70% |
| 7 | [Grok 4.6](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±6.2 · 799 reported samples | xAI | 56.40% |
| 8 | [Opus 5.5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · medium reasoning · Published margin ±6.3 · 801 reported samples | Anthropic | 55.90% |
| 9 | [Haiku 5.5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.5 · 800 reported samples | Anthropic | 55.10% |
| 10 | [MiMo V2.6 Pro](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · auto reasoning · Published margin ±6.2 · 793 reported samples | Xiaomi | 54.20% |
| 11 | [Grok 4.5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±6.4 · 800 reported samples | xAI | 53.60% |
| 12 | [Grok 4.7](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · xhigh reasoning · Published margin ±6.3 · 800 reported samples | xAI | 53.40% |
| 13 | [Qwen3.8-Max](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · xhigh reasoning · Published margin ±6.4 · 799 reported samples | Alibaba | 52.30% |
| 14 | [DeepSeek-V4.1-Flash](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.3 · 800 reported samples | DeepSeek | 51.90% |
| 15 | [Qwen 3.8 27B](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · xhigh reasoning · Published margin ±6.2 · 800 reported samples | Alibaba | 50.70% |
| 16 | [Gemini 4 Argon](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±6.6 · 800 reported samples | Google | 49.20% |
| 17 | [GPT-6 Astra](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.3 · 1,600 reported samples | OpenAI | 48.60% |
| 18 | [Kimi K3](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.2 · 800 reported samples | Kimi | 48.00% |
| 19 | [GPT-5.6 Sol](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.4 · 800 reported samples | OpenAI | 48.00% |
| 20 | [Opus 4.7](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.5 · 800 reported samples | Anthropic | 47.60% |
| 21 | [GPT-6.1 Sol](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.4 · 800 reported samples | OpenAI | 46.90% |
| 22 | [Sonnet 5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.4 · 800 reported samples | Anthropic | 46.40% |
| 23 | [GLM-5.3-Flash](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.1 · 800 reported samples | Zhipu | 46.10% |
| 24 | [DeepSeek-V4-Pro-0813](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.1 · 1,570 reported samples | DeepSeek | 45.90% |
| 25 | [GPT-5.6 Sol](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · xhigh reasoning · Published margin ±6.2 · 800 reported samples | OpenAI | 45.80% |
| 26 | [GLM-5.3](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±5.9 · 800 reported samples | Zhipu | 45.60% |
| 27 | [GPT-6 Sol](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.3 · 800 reported samples | OpenAI | 45.00% |
| 28 | [Sonnet 5.5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · medium reasoning · Published margin ±6.1 · 800 reported samples | Anthropic | 44.00% |
| 29 | [Opus 4.8](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.1 · 800 reported samples | Anthropic | 43.90% |
| 30 | [Gemini 3.7 Flash](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±6.4 · 800 reported samples | Google | 42.00% |
| 31 | [Opus 4.6](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.2 · 795 reported samples | Anthropic | 41.80% |
| 32 | [Muse Spark 1.2](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · xhigh reasoning · Published margin ±6.2 · 800 reported samples | Meta | 40.50% |
| 33 | [Kimi K2.7 Code](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±5.9 · 800 reported samples | Kimi | 40.30% |
| 34 | [Gemini 3.6 Flash](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±6.2 · 800 reported samples | Google | 39.40% |
| 35 | [Muse Spark 1.3](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.4 · 1,592 reported samples | Meta | 39.10% |
| 36 | [Muse Spark 1.1](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · xhigh reasoning · Published margin ±6.1 · 656 reported samples | Meta | 38.90% |
| 37 | [GPT-6 Luna](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.1 · 800 reported samples | OpenAI | 38.80% |
| 38 | [GLM-5.2](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±6.1 · 800 reported samples | Zhipu | 37.40% |
| 39 | [GPT-5.5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · xhigh reasoning · Published margin ±5.8 · 1,600 reported samples | OpenAI | 36.60% |
| 40 | [Gemini 3.8 Flash](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±5.9 · 1,595 reported samples | Google | 36.40% |
| 41 | [Muse Spark 1.3](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · xhigh reasoning · Published margin ±6.4 · 796 reported samples | Meta | 36.40% |
| 42 | [Gemini 3.5 Flash](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±6 · 800 reported samples | Google | 36.10% |
| 43 | [MiniMax-M3](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±5.6 · 1,600 reported samples | MiniMax | 34.60% |
| 44 | [GLM-5.1](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · Published margin ±5.7 · 800 reported samples | Zhipu | 34.50% |
| 45 | [Sonnet 4.6](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±6.1 · 800 reported samples | Anthropic | 34.50% |
| 46 | [Gemini 3.1 Pro](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±5.8 · 800 reported samples | Google | 33.90% |
| 47 | [DeepSeek-V4-Pro](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±5.7 · 800 reported samples | DeepSeek | 33.80% |
| 48 | [GPT-5.6 Luna](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±5.6 · 800 reported samples | OpenAI | 32.10% |
| 49 | [Gemini 3.5 Flash Lite](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±5.6 · 800 reported samples | Google | 31.20% |
| 50 | [GPT-5.6 Terra](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · max reasoning · Published margin ±5.6 · 800 reported samples | OpenAI | 31.10% |
| 51 | [Qwen3.5](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · Published margin ±5.1 · 2,399 reported samples | Alibaba | 27.90% |
| 52 | [Inkling](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±5.2 · 800 reported samples | Thinking Machines | 26.10% |
| 53 | [DeepSeek-V3.2](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · Published margin ±5.1 · 800 reported samples | DeepSeek | 25.60% |
| 54 | [MiniMax-M2.7](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±4.6 · 800 reported samples | MiniMax | 21.50% |
| 55 | [Kimi K2 (Thinking)](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±4.2 · 800 reported samples | Kimi | 17.40% |
| 56 | [Nemotron 3 Ultra](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±3.6 · 1,579 reported samples | NVIDIA | 15.20% |
| 57 | [GPT-OSS-120B](https://www.mercor.com/apex/apex-swe-leaderboard/) | Terminus-2 · high reasoning · Published margin ±1.6 · 800 reported samples | OpenAI | 3.30% |

## FAQ

### What does APEX-SWE (Terminus-2) (Mercor) measure?

Software integration and production-failure diagnosis with the Terminus-2 harness. This table shows Mercor-run configurations for reference and is excluded from model rankings.

### Which model leads the published APEX-SWE (Terminus-2) (Mercor) snapshot?

Opus 5.5 currently leads the published APEX-SWE (Terminus-2) (Mercor) snapshot with a score of 67.60%.

### How many models are evaluated on APEX-SWE (Terminus-2) (Mercor)?

The October 15, 2026 capture contains 57 source configurations.

### Does APEX-SWE (Terminus-2) (Mercor) affect BenchLM's overall score?

Not directly. APEX-SWE (Terminus-2) (Mercor) is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.