Dataset Viewer
Auto-converted to Parquet Duplicate
date
large_stringlengths
8
8
channels
int64
4.86k
50.9k
nodes
int64
2.32k
13.1k
cap_btc
float64
746
5.84k
fresh14_share
float64
0
0.98
epoch
large_stringclasses
3 values
status
large_stringclasses
3 values
20230508
42,321
8,335
3,957.041222
0.966612
A
ok
20230808
42,318
8,702
3,610.647962
0.957418
A
ok
20230810
42,394
8,715
3,628.350519
0.956598
A
ok
20231020
44,565
8,537
4,675.005125
0.963963
A
ok
20231022
44,006
8,499
4,620.827511
0.961732
A
ok
20231023
43,900
8,493
4,540.112296
0.96082
A
ok
20231030
43,890
8,403
4,597.794574
0.95753
A
ok
20231101
43,856
8,393
4,625.510625
0.959709
A
ok
20231102
43,924
8,399
4,639.760335
0.960067
A
ok
20231103
43,934
8,398
4,643.791488
0.960054
A
ok
20231106
43,977
8,404
4,656.775149
0.961184
A
ok
20231107
43,970
8,411
4,639.873861
0.962156
A
ok
20231108
43,970
8,422
4,644.720995
0.961906
A
ok
20231109
43,920
8,428
4,617.924868
0.962204
A
ok
20231111
43,896
8,428
4,630.36338
0.960908
A
ok
20231113
43,867
8,437
4,636.993514
0.963047
A
ok
20231114
43,835
8,434
4,637.883941
0.961059
A
ok
20231115
43,781
8,430
4,648.316219
0.962084
A
ok
20231116
43,771
8,432
4,650.65143
0.960659
A
ok
20231117
43,738
8,439
4,638.712323
0.959555
A
ok
20231121
43,706
8,459
4,685.866075
0.959205
A
ok
20231122
43,703
8,462
4,689.96853
0.957097
A
ok
20231125
43,646
8,450
4,688.395721
0.949228
A
ok
20231126
43,640
8,445
4,688.344826
0.945555
A
ok
20231127
43,625
8,440
4,687.892094
0.939484
A
ok
20231130
43,017
8,440
4,458.38772
0.926936
A
ok
20231204
42,981
8,443
4,453.01401
0.930876
A
ok
20231205
42,771
8,418
4,443.55942
0.925393
A
ok
20231208
42,724
8,422
4,448.463567
0.937506
A
ok
20231210
42,655
8,421
4,465.949159
0.944883
A
ok
20231211
42,635
8,433
4,444.255831
0.947578
A
ok
20231212
42,623
8,448
4,430.583938
0.949487
A
ok
20231213
42,643
8,452
4,434.141618
0.950379
A
ok
20231215
42,645
8,459
4,443.431535
0.952327
A
ok
20231217
42,559
8,460
4,429.07935
0.954064
A
ok
20231219
42,564
8,474
4,427.518516
0.953623
A
ok
20231220
42,575
8,485
4,420.221961
0.952813
A
ok
20231221
42,549
8,474
4,427.000905
0.951186
A
ok
20231223
42,512
8,487
4,384.922135
0.950273
A
ok
20231225
42,507
8,485
4,385.259897
0.950808
A
ok
20231227
42,539
8,495
4,396.600268
0.949388
A
ok
20231228
42,484
8,504
4,429.779891
0.948922
A
ok
20231231
42,393
8,517
4,410.082041
0.9461
A
ok
20240102
42,200
8,530
4,369.095896
0.948081
A
ok
20240103
42,156
8,562
4,377.819122
0.948643
A
ok
20240104
42,135
8,563
4,393.877311
0.947455
A
ok
20240105
42,159
8,582
4,400.213149
0.949121
A
ok
20240106
42,095
8,568
4,406.298764
0.949305
A
ok
20240108
42,129
8,569
4,436.102357
0.949607
A
ok
20240109
42,102
8,563
4,462.899997
0.949955
A
ok
20240110
42,137
8,570
4,459.564202
0.950803
A
ok
20240111
41,841
8,573
4,239.76648
0.952582
A
ok
20240112
41,836
8,571
4,237.156649
0.951884
A
ok
20240113
41,764
8,552
4,232.022549
0.951513
A
ok
20240114
41,769
8,561
4,238.657783
0.951902
A
ok
20240115
41,768
8,560
4,240.386643
0.949938
A
ok
20240117
41,745
8,573
4,291.979717
0.95106
A
ok
20240118
41,686
8,580
4,287.030063
0.951782
A
ok
20240119
41,696
8,583
4,300.537652
0.950955
A
ok
20240122
41,776
8,602
4,333.069216
0.950689
A
ok
20240123
41,695
8,608
4,286.223706
0.950042
A
ok
20240124
41,688
8,609
4,283.39938
0.949818
A
ok
20240125
41,623
8,607
4,285.238142
0.949715
A
ok
20240126
41,605
8,616
4,305.948681
0.949501
A
ok
20240127
41,637
8,623
4,303.231054
0.949636
A
ok
20240129
41,761
8,641
4,336.615655
0.948373
A
ok
20240131
41,800
8,642
4,369.787437
0.947392
A
ok
20240201
41,823
8,645
4,390.286008
0.947899
A
ok
20240203
41,617
8,647
4,392.939111
0.946416
A
ok
20240204
41,595
8,647
4,389.253634
0.944633
A
ok
20240205
41,585
8,644
4,389.073309
0.942335
A
ok
20240207
41,580
8,641
4,388.960309
0.936003
A
ok
20240211
41,580
8,641
4,388.960309
0.91746
A
ok
20240212
40,014
8,376
4,216.513534
0.930224
A
ok
20240213
40,028
8,360
4,233.881396
0.932123
A
ok
20240214
40,039
8,363
4,241.456556
0.931242
A
ok
20240215
39,916
8,347
4,264.707287
0.931957
A
ok
20240219
39,273
8,245
4,129.439646
0.939577
A
ok
20240220
39,216
8,241
4,131.669238
0.943926
A
ok
20240222
39,233
8,237
4,182.586479
0.949609
A
ok
20240223
39,174
8,246
4,089.199242
0.950733
A
ok
20240224
39,225
8,247
4,071.715946
0.952913
A
ok
20240226
39,048
8,204
4,082.33939
0.955721
A
ok
20240227
39,088
8,210
4,098.118958
0.954897
A
ok
20240229
38,971
8,226
4,027.281702
0.955736
A
ok
20240304
39,049
8,243
4,066.564518
0.955594
A
ok
20240305
38,974
8,243
4,069.701309
0.955688
A
ok
20240306
38,907
8,238
4,059.831265
0.955047
A
ok
20240311
38,660
8,205
4,045.2192
0.95507
A
ok
20240312
38,587
8,196
4,068.817789
0.955166
A
ok
20240313
38,365
8,159
4,065.04136
0.954985
A
ok
20240315
38,244
8,151
4,090.542835
0.95432
A
ok
20240316
38,232
8,148
4,090.285592
0.951664
A
ok
20240317
38,223
8,142
4,090.094112
0.949062
A
ok
20240318
38,223
8,142
4,090.094112
0.946132
A
ok
20240319
38,223
8,142
4,090.094112
0.942731
A
ok
20240320
38,223
8,142
4,090.094112
0.939461
A
ok
20240321
38,223
8,142
4,090.094112
0.93483
A
ok
20240324
37,728
8,074
4,044.797015
0.927295
A
ok
20240325
37,709
8,068
4,049.030626
0.925588
A
ok
End of preview. Expand in Data Studio

Lightning Network Daily Gossip Snapshots (describegraph), 2023–2026

Open daily Parquet snapshots of the public Bitcoin Lightning Network gossip graph captured from our LND node (describegraph), curated by Bitcoin Data Labs (501(c)(3)).


Coverage & Quality Manifest

  • Span: 20230508 through 20261115 (847 quality-checked ok days included; 899 total raw captures documented in _quality.parquet / quality.parquet: 847 ok, 51 partial, 1 stale).
  • Reproducing arXiv:2610.20952, arXiv:2610.30350, and arXiv:2611.16311: Our research series truncates snapshots at fixed cutoffs (DATA_CUTOFF = "20261004" for Papers 1–2: 805 ok days; 2026-10 block 974,041 for Paper 3). Using any companion repo, extract the Zenodo yearly archives into one gossip_daily/ directory alongside _quality.parquet and run:
    export LN_PARQUET_DIR=/path/to/parquet_root   # containing gossip_daily/{nodes,policies,_quality.parquet}
    python src/01_client_census.py
    

Vantage Point Epochs (Important for Time-Series Analysis)

Our gossip vantage node was replaced in mid-2024, which changes baseline channel and node visibility. Compare levels strictly within an epoch (or filter epoch == 'B' in quality.parquet for a homogeneous post-replacement panel):

Epoch / Window Dates ok Days Notes
Epoch A 2023-05-08 – 2024-04-27 132 Prior vantage node (~37–44k channels, ~8.0–8.6k nodes with channels).
gap (partial, excluded) 2024-05-07 – 2024-06-30 0 (51 partial) Partial captures (4.9k–8.8k channels, except 2024-06-10). Exact vantage jump occurred 2024-06-20 -> 2024-06-21 (44,025 -> 50,867 channels; 12,254 -> 13,037 nodes).
Epoch B 2024-07-01 – 20261115 715 Post-replacement vantage point (~41k median channels in 2024–early 2026; ~26.5k channels by Nov 2026). Includes documented missing "wallet locked" periods (2024-07-24–2024-08-05, 2025-03-27–2025-05-18) and 1 stale day (2024-08-06, excluded).

Responsible Use & Data Ethics

  • All fields come solely from publicly announced Lightning gossip (describegraph).
  • In accordance with Bitcoin Data Labs' responsible-disclosure standard (bitcoindatalabs.org), published security analyses should remain aggregates only (no naming, ranking, or labeling of individual nodes by software vulnerability or exploit exposure).
  • License: Compilation, schemas, and quality manifest are licensed under CC BY 4.0. Credit: "Source: Bitcoin Data Labs (bitcoindatalabs.org)".

Quick Usage Examples

Hugging Face + DuckDB (Zero Full Download)

SELECT count(*), sum(capacity) / 2e8 AS total_cap_btc
FROM read_parquet('hf://datasets/BitcoindataLabs/ln-gossip-daily/data/policies/year=2026/date=20261004.parquet');

Python (huggingface_hub + pandas)

from huggingface_hub import hf_hub_download
import pandas as pd

quality = pd.read_parquet(
    hf_hub_download("BitcoindataLabs/ln-gossip-daily", "data/quality.parquet", repo_type="dataset")
)
epoch_b_dates = quality.query("epoch == 'B' and status == 'ok'")["date"].tolist()

Query Pre-Aggregated Benchmarks via MCP (bitcoindatalabs-mcp)

Need aggregate paper metrics, fee-spike close ratios, release-lag statistics, or weekly PlebRank centrality without scanning all 5.7 GiB locally? Run our MIT-licensed MCP server in Claude, Cursor, or VS Code (DOI: 10.5281/zenodo.24285334):

{
  "mcpServers": {
    "bitcoindatalabs-mcp": {
      "command": "uvx",
      "args": ["bitcoindatalabs-mcp"]
    }
  }
}

How to Cite

@misc{kumar2026lngossipdaily,
  author       = {Kumar, Saurabh},
  title        = {Lightning Network Daily Gossip Snapshots (describegraph), 2023--2026},
  year         = {2026},
  month        = nov,
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.24292148},
  url          = {https://doi.org/10.5281/zenodo.24292148}
}

@misc{kumar2026clientdiversity,
  author        = {Kumar, Saurabh},
  title         = {Does Client Diversity Protect the Lightning Network? Evidence from the 2026 Core Lightning Shutdown},
  year          = {2026},
  month         = oct,
  eprint        = {2610.20952},
  archivePrefix = {arXiv},
  primaryClass  = {cs.CR},
  doi           = {10.48550/arXiv.2610.20952},
  url           = {https://arxiv.org/abs/2610.20952}
}

@misc{kumar2026upgradedynamics,
  author        = {Kumar, Saurabh},
  title         = {Who Patches the Lightning Network? Evidence from Release Lines in Gossip, 2023--2026},
  year          = {2026},
  month         = oct,
  eprint        = {2610.30350},
  archivePrefix = {arXiv},
  primaryClass  = {cs.CR},
  doi           = {10.48550/arXiv.2610.30350},
  url           = {https://arxiv.org/abs/2610.30350}
}

@misc{kumar2026forceclosecosts,
  author        = {Kumar, Saurabh},
  title         = {What Does It Cost to Force-Close a Lightning Channel? Evidence from 974,000 Channel Closes, 2018--2026},
  year          = {2026},
  month         = nov,
  eprint        = {2611.16311},
  archivePrefix = {arXiv},
  primaryClass  = {cs.CR},
  doi           = {10.48550/arXiv.2611.16311},
  url           = {https://arxiv.org/abs/2611.16311}
}
Downloads last month
63

Papers for BitcoindataLabs/ln-gossip-daily