# SWE Atlas - Codebase QnA — Mercor evaluation (SWE Atlas - Codebase QnA (Mercor))

> Answering questions about software repositories with an agent. This table shows Mercor-run configurations for reference and is excluded from model rankings.

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

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

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

We captured Mercor's SWE Atlas - Codebase QnA results on October 15, 2026. The table preserves 41 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.

The headline is the source page's selected Pass@1 metric. The task set, harness, tools, and grader belong to this Mercor run and may differ from another evaluator's results.

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 llm-as-a-judge (lmaaj) over software quality rubrics and a judge of Claude Opus 4.5. Its run-count label is k = 3. The linked result page takes precedence where its setup differs, as noted above.

- [Mercor SWE Atlas - Codebase QnA results](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/)
- [Mercor APEX benchmark index](https://www.mercor.com/apex/)
- [Mercor methodology](https://www.mercor.com/apex/methodology/)
- [Source paper](https://arxiv.org/abs/2605.08366)
- [Upstream evaluation code](https://github.com/scaleapi/SWE-Atlas)
- [Upstream dataset](https://github.com/scaleapi/SWE-Atlas)

## About SWE Atlas - Codebase QnA (Mercor)

- Tasks: 124 source-reported tasks
- Format: Pass@1
- Difficulty: Source-specific evaluation
- Paper: [SWE Atlas - Codebase QnA evaluation results](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-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.

SWE Atlas - Codebase QnA (Mercor) is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (41 configurations)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [DeepSeek-V4.1-Flash](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.4 · 372 reported samples | DeepSeek | 53.50% |
| 2 | [Kimi K3](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.1 · 372 reported samples | Kimi | 52.70% |
| 3 | [Qwen3.8-Max](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · xhigh reasoning · Published margin ±7.7 · 372 reported samples | Alibaba | 50.50% |
| 4 | [Opus 5](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.7 · 372 reported samples | Anthropic | 50.30% |
| 5 | [Sonnet 5.5](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.8 · 372 reported samples | Anthropic | 50.30% |
| 6 | [Grok 4.6](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · xhigh reasoning · Published margin ±7.4 · 372 reported samples | xAI | 50.30% |
| 7 | [Opus 5.5](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.9 · 372 reported samples | Anthropic | 48.90% |
| 8 | [GPT-6 Astra](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · xhigh reasoning · Published margin ±7.9 · 372 reported samples | OpenAI | 48.90% |
| 9 | [Haiku 5.5](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.4 · 372 reported samples | Anthropic | 48.70% |
| 10 | [GPT-6.1 Sol](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.7 · 372 reported samples | OpenAI | 48.40% |
| 11 | [GPT-6 Astra](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.5 · 372 reported samples | OpenAI | 47.80% |
| 12 | [Opus 4.7](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.4 · 372 reported samples | Anthropic | 47.00% |
| 13 | [Grok 4.7](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · xhigh reasoning · Published margin ±7 · 372 reported samples | xAI | 47.00% |
| 14 | [GLM-5.2](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.5 · 372 reported samples | Zhipu | 46.00% |
| 15 | [GPT-5.6 Luna](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.7 · 372 reported samples | OpenAI | 46.00% |
| 16 | [GPT-5.5](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · xhigh reasoning · Published margin ±7.5 · 372 reported samples | OpenAI | 45.20% |
| 17 | [GPT-5.6 Terra](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.5 · 372 reported samples | OpenAI | 45.20% |
| 18 | [DeepSeek-V4-Flash](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.3 · 372 reported samples | DeepSeek | 44.90% |
| 19 | [Opus 4.8](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.5 · 372 reported samples | Anthropic | 44.40% |
| 20 | [GPT-6 Sol](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.7 · 372 reported samples | OpenAI | 44.10% |
| 21 | [Fable 5.1](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.9 · 372 reported samples | Anthropic | 43.30% |
| 22 | [Fable 5.1](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±7.9 · 372 reported samples | Anthropic | 42.70% |
| 23 | [MiMo V2.6 Pro](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · auto reasoning · Published margin ±7.3 · 372 reported samples | Xiaomi | 42.70% |
| 24 | [GLM-5.3](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7 · 371 reported samples | Zhipu | 42.50% |
| 25 | [Grok 4.5](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±7.1 · 372 reported samples | xAI | 41.90% |
| 26 | [Gemini 3.8 Flash](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±7.5 · 372 reported samples | Google | 40.60% |
| 27 | [Sonnet 5](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.3 · 372 reported samples | Anthropic | 39.50% |
| 28 | [MiniMax-M3](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±7 · 372 reported samples | MiniMax | 37.90% |
| 29 | [GPT-5.4](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · xhigh reasoning · Published margin ±7 · 372 reported samples | OpenAI | 37.10% |
| 30 | [Sonnet 4.6](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±7.3 · 372 reported samples | Anthropic | 35.50% |
| 31 | [GPT-6 Luna](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.7 · 372 reported samples | OpenAI | 33.90% |
| 32 | [Kimi K2.7 Code](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±7.1 · 372 reported samples | Kimi | 33.10% |
| 33 | [Fable 5](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±7.3 · 372 reported samples | Anthropic | 33.10% |
| 34 | [Gemini 3.6 Flash](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±7.1 · 372 reported samples | Google | 32.00% |
| 35 | [GPT-5.6 Sol (Pro)](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±6.8 · 365 reported samples | OpenAI | 30.90% |
| 36 | [DeepSeek-V4-Pro](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · max reasoning · Published margin ±6.6 · 372 reported samples | DeepSeek | 25.00% |
| 37 | [Inkling](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±6 · 372 reported samples | Thinking Machines | 21.20% |
| 38 | [Gemini 3.5 Flash](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±5.8 · 372 reported samples | Google | 19.60% |
| 39 | [Gemini 3.1 Pro](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · high reasoning · Published margin ±6 · 372 reported samples | Google | 18.00% |
| 40 | [DeepSeek-V3.2](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · Published margin ±3.4 · 372 reported samples | DeepSeek | 5.90% |
| 41 | [Muse Spark 1.1](https://www.mercor.com/apex/oss-benchmarks/oss-swe-atlas-leaderboard/) | mini-swe-agent (250 max steps, 1 hour time limit) · xhigh reasoning · Published margin ±2.6 · 370 reported samples | Meta | 3.90% |

## FAQ

### What does SWE Atlas - Codebase QnA (Mercor) measure?

Answering questions about software repositories with an agent. This table shows Mercor-run configurations for reference and is excluded from model rankings.

### Which model leads the published SWE Atlas - Codebase QnA (Mercor) snapshot?

DeepSeek-V4.1-Flash currently leads the published SWE Atlas - Codebase QnA (Mercor) snapshot with a score of 53.50%.

### How many models are evaluated on SWE Atlas - Codebase QnA (Mercor)?

The October 15, 2026 capture contains 41 source configurations.

### Does SWE Atlas - Codebase QnA (Mercor) affect BenchLM's overall score?

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