# SciCode Main With Background (Kaggle task v3) (SciCode Main With Background (Kaggle))

> A separate Kaggle SciCode track for main research problems with supplied scientific background.

Canonical page: https://benchlm.ai/benchmarks/scicode-main-with-background-kaggle

- Category: [Coding](/coding)
- Last updated: 2026-10-20 source capture

Evaluation results published by [Kaggle Benchmarks](https://www.kaggle.com/benchmarks/open-benchmarks/scicode). undefined

This Kaggle SciCode track tests main research problems with supplied scientific background. The four tracks retain separate scores because supplying background information and evaluating subproblems change the task.

We preserve Kaggle task version 3, source configuration names, evaluation dates, and published intervals. The source does not state an interval confidence level here. These rows remain separate from the core SciCode, Artificial Analysis, and Mercor evaluation protocols.

These are published source configurations. Compare the task version and disclosed settings before treating a score difference as a model difference.

- [Kaggle leaderboard and methodology](https://www.kaggle.com/benchmarks/open-benchmarks/scicode)
- [Published leaderboard data](https://www.kaggle.com/api/v1/benchmarks/open-benchmarks/scicode/leaderboard)
- [SciCode project and original methodology](https://scicode-bench.github.io/)
- [SciCode paper](https://arxiv.org/abs/2407.13168)
- [Apache 2.0 license shown by Kaggle](https://www.apache.org/licenses/LICENSE-2.0)

## About SciCode Main With Background (Kaggle)

- Year: 2024
- Tasks: Scientific research main problems
- Format: Published Kaggle leaderboard aggregates
- Paper: [SciCode: A Research Coding Benchmark Curated by Scientists](https://arxiv.org/abs/2407.13168)

This track reports problem success for main research problems with supplied scientific background. We show the published percentage and keep it separate from the other three Kaggle tracks and other SciCode evaluation providers.

SciCode Main With Background (Kaggle) is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (62 configurations)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Gemini 3.1 Pro Preview](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3 | — | 30.0% |
| 2 | [GPT-5.6 Sol](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-07-22 | — | 23.1% |
| 3 | [Claude Sonnet 5](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-07-23 | — | 21.7% |
| 4 | [Claude Opus 5.5](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-09-24 | — | 21.5% |
| 5 | [Gemini 3.7 Flash](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-08-28 | — | 21.5% |
| 6 | [Grok 4.5](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-07-22 | — | 21.5% |
| 7 | [Claude Opus 5](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-07-28 | — | 20.0% |
| 8 | [Gemini 3.8 Flash](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-09-06 | — | 20.0% |
| 9 | [Claude Sonnet 5.5](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-09-30 | — | 18.5% |
| 10 | [GPT-6 Astra](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-09-14 | — | 18.5% |
| 11 | [Gemini 4 Argon](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-10-12 | — | 18.5% |
| 12 | [GPT-5.6 Terra](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-07-22 | — | 18.5% |
| 13 | [GPT-6.1 Sol](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-10-05 | — | 18.5% |
| 14 | [Gemini 3.5 Flash](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3 | — | 18.5% |
| 15 | [Gemini 3 Flash Preview](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-07-06 | — | 16.9% |
| 16 | [Claude Opus 4.8](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3 | — | 16.9% |
| 17 | [Gemini 3.6 Flash](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-07-22 | — | 16.9% |
| 18 | [Gemini 2.5 Pro Preview](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-06-06 | — | 15.4% |
| 19 | [Grok 3 Mini](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-26 | — | 13.8% |
| 20 | [GPT-5.6 Luna](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-07-22 | — | 12.3% |
| 21 | [Gemini 3.5 Flash-Lite](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-07-22 | — | 12.3% |
| 22 | [o3 mini](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 12.3% |
| 23 | [Claude Opus 4](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-23 | — | 12.3% |
| 24 | [Claude Haiku 5.5](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2026-10-09 | — | 10.8% |
| 25 | [Claude Sonnet 4](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-23 | — | 10.8% |
| 26 | [o1](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 10.8% |
| 27 | [DeepSeek-R1](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-06-04 | — | 10.8% |
| 28 | [DeepSeek-V3](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 10.8% |
| 29 | [Claude 3.5 Sonnet](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 10.8% |
| 30 | [Gemma 4 31B](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3 | — | 9.8% |
| 31 | [o4 mini](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 9.2% |
| 32 | [o3](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 9.2% |
| 33 | [Gemini 1.5 Pro](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 9.2% |
| 34 | [GPT-4.1](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 9.2% |
| 35 | [Claude Opus 4.1](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-08-26 | — | 9.2% |
| 36 | [deepseek-v3.1](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-08-27 | — | 7.7% |
| 37 | [Claude 3.7 Sonnet](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 7.7% |
| 38 | [Gemini 2.5 Flash Preview](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-23 | — | 7.7% |
| 39 | [GPT-4o](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 7.7% |
| 40 | [gpt-oss-120b](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-08-15 | — | 6.2% |
| 41 | [Qwen 3 235B A22B Instruct 2506](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-09-03 | — | 6.2% |
| 42 | [Gemini 3.1 Flash-Lite Preview](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3 | — | 6.2% |
| 43 | [Mistral Large 2](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 6.2% |
| 44 | [Grok 3](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-26 | — | 6.2% |
| 45 | [gpt-oss-20b](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-08-15 | — | 6.2% |
| 46 | [Gemini 1.5 Flash 8B](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 4.6% |
| 47 | [o1 mini](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 3.1% |
| 48 | [Gemini 2.0 Flash](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 3.1% |
| 49 | [Gemini 1.5 Flash](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 3.1% |
| 50 | [Grok 2](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 3.1% |
| 51 | [Codestral](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 3.1% |
| 52 | [Mixtral 8x22B](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 3.1% |
| 53 | [Granite 4.0 Small](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-09-30 | — | 1.5% |
| 54 | [GPT-4o mini](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 0.0% |
| 55 | [Claude 3.5 Haiku](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 0.0% |
| 56 | [GPT-3.5 Turbo](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 0.0% |
| 57 | [Ministral 3B](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 0.0% |
| 58 | [Ministral 8B](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 0.0% |
| 59 | [Gemma 3 4B](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 0.0% |
| 60 | [Gemma 3 27B](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 0.0% |
| 61 | [Gemma 3 12B](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3; Evaluated 2025-05-22 | — | 0.0% |
| 62 | [GPT-5](https://www.kaggle.com/benchmarks/open-benchmarks/scicode) | Task version 3 | — | 0.0% |

## FAQ

### What does SciCode Main With Background (Kaggle) measure?

A separate Kaggle SciCode track for main research problems with supplied scientific background.

### Which source row has the highest reported SciCode Main With Background (Kaggle) value?

Gemini 3.1 Pro Preview has the highest reported SciCode Main With Background (Kaggle) value at 30.0%. These are published source configurations. Compare the task version and disclosed settings before treating a score difference as a model difference.

### How many models are evaluated on SciCode Main With Background (Kaggle)?

The 2026-10-20 source capture contains 62 AI models.

### Does SciCode Main With Background (Kaggle) affect BenchLM's overall score?

Not directly. SciCode Main With Background (Kaggle) is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.