aqora / Colour Codes Reach Surface Code Performance using Vibe Decoding
Public
323 rows, 15 columns
No selectionPageof 33
No selection
SELECT * FROM dataVibeLSD Logical Error Rates — Color-Code Decoding Benchmarks
Overview
This dataset contains logical error rates obtained from large-scale simulations of VibeLSD decoding across a wide range of color-code circuits, noise models, and ensemble sizes. It accompanies the study “Colour Codes Reach Surface Code Performance using Vibe Decoding” (arXiv:2508.15743). The original data were produced in a format compatible with Sinter, enabling direct comparison of decoders and circuit families. This Aqora release preserves the scientific content exactly while providing a columnar representation suitable for scalable analytics, plotting, and ML workflows.Scientific Context
The dataset reports logical error rates measured for different color-code circuit constructions, including:- midout circuits
- superdense circuits
- bell-flagged circuits
- mxyz circuits
- tri-optimal circuits
Dataset Contents
- Rows: 323 decoder benchmark points
- Granularity: one row per
(circuit, noise point, decoder configuration)
Parquet Schema
Each row corresponds to one logical-error-rate estimate.| Column | Type | Description |
|---|---|---|
csv_filename | String | Original CSV file name. |
circuit_family | String | Circuit family (midout, superdense, bellflag, mxyz, trioptimal). |
noise_model | String | Noise model identifier (e.g. uniform, si1000). |
ensemble_size | Int32 | Number of parallel decoder instances. |
shots | Int64 | Total number of Monte-Carlo samples. |
errors | Int64 | Number of logical failures observed. |
discards | Int64 | Number of discarded shots. |
seconds | Float64 | Reported wall-clock runtime. |
decoder | String | Decoder name and configuration (e.g. VLSD-32). |
strong_id | String | Unique identifier encoding circuit and resolution. |
custom_counts | String (nullable) | Optional decoder-specific counters. |
json_metadata_raw | String | Raw JSON metadata string from the CSV. |
meta | Struct | Parsed metadata fields (see below). |
logical_error_rate | Float64 | Logical error rate (errors / shots). |
discard_rate | Float64 | Fraction of discarded shots (discards / shots). |
Column availability depends on the original CSV; missing values are stored as nulls.
Loading Examples
Polars
import polars as pl
from aqora_cli.pyarrow import dataset
df = pl.scan_pyarrow_dataset(
dataset("aqora/vibe-decoding-logical-error-rates", "v0.0.0")
).collect()
print(df.head())
Pandas
import pandas as pd
df = pd.read_parquet(
"aqora://aqora/vibe-decoding-logical-error-rates/v0.0.0"
)
print(df.head())
Citation
If you use this dataset, please cite the original work:Colour Codes Reach Surface Code Performance using Vibe Decoding, 2025. Stergios Koutsioumpas, Tamas Noszko, Hasan Sayginel, Mark Webster, Joschka Roffe arXiv:2508.15743 (2025). https://arxiv.org/abs/2508.15743