MLS-Bench
MLS-Bench - machine learning systems agent benchmark (e.g. safe-RL constraint-handling mechanism design). Scores are measured over the MLS-Bench-Tasks dataset (https://huggingface.co/datasets/Bohan22/MLS-Bench-Tasks); official site: https://mls-bench.com/.
Category
Agentic Coding
Score type
percent
Scale
100%
higher is better
Models tested
3
What is MLS-Bench?
MLS-Bench is a benchmark for how well a model works as an autonomous coding agent. The official description says: "MLS-Bench - machine learning systems agent benchmark (e.g". Below we describe in simple words what benchmarks of this type generally examine.
What does it test?
MLS-Bench measures how well a model works as an autonomous coding agent. In detail, it checks:
- fixing real, open bugs from real software repositories
- planning multi-step changes, running commands and editing files by itself
- passing the project's test suite after its changes
- handling whole repositories, not just small isolated code snippets
How to read the score
The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.
Official methodology note: MLS-Bench - machine learning systems agent benchmark (e.g. safe-RL constraint-handling mechanism design). Scores are measured over the MLS-Bench-Tasks dataset (https://huggingface.co/datasets/Bohan22/MLS-Bench-Tasks); official site: https://mls-bench.com/.
Frequently asked questions
What is the MLS-Bench benchmark?
MLS-Bench is a benchmark for how well a model works as an autonomous coding agent. The official description says: 'MLS-Bench - machine learning systems agent benchmark (e.g'. Below we describe in simple words what benchmarks of this type generally examine.
What does MLS-Bench test?
MLS-Bench mainly measures how well a model works as an autonomous coding agent. Concretely, it checks: fixing real, open bugs from real software repositories; planning multi-step changes, running commands and editing files by itself; passing the project's test suite after its changes.
How is the MLS-Bench score calculated?
The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.
Where do I get MLS-Bench data and scores?
The official source is https://huggingface.co/datasets/Bohan22/MLS-Bench-Tasks. The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/mls-bench/scores) or pull it directly from the Open LLM Radar API with any HTTP client.
Benchmark data & downloads
Where to get, inspect and reuse the data behind this benchmark.
Official benchmark page / paper
The benchmark maintainer's website, paper or leaderboard
Benchmark dataset
The question/task data itself — download and evaluate locally
Benchmark metadata (JSON)
This page's benchmark record from the Open LLM Radar API
All model scores (JSON download)
Full score table as JSON; paginate with limit/offset params
Open LLM Radar API docs
Swagger UI to explore and reuse all radar API endpoints
Model scores 3 models tested
| # | Model | Score | Date | Source | Verified |
|---|---|---|---|---|---|
| 1 |
DeepSeek-V4.1-Flash-1017
DeepSeek
|
44.3%
|
— | Report | Unverified |
| 2 |
Sage-S2.1-Flash
Shanghai AI Laboratory
|
42%
|
— | Report | Verified |
| 3 |
GLM-5.3-Flash
Zhipu AI
|
39.9%
|
— | Report | Unverified |