huggingface/agent-usage · Datasets at Hugging Face
Log In<br>Sign Up
User-Agent token identifying the… See the full description on the dataset page: https://huggingface.co/datasets/huggingface/agent-usage.","downloads":0,"downloadsAllTime":0,"id":"huggingface/agent-usage","isLikedByUser":false,"lastModified":"2026-07-02T15:14:55.000Z","likes":1,"datasetsServerInfo":{"viewer":"viewer","numRows":1372,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet"],"modalities":["tabular","text"]},"discussionsDisabled":false,"discussionsSorting":"recently-created","repoType":"dataset","private":false,"gated":false,"tags":["size_categories:1K
Dataset Viewer
Auto-converted to Parquet API Embed Duplicate Data Studio<br>Subset (2) monthly · 53 rows<br>monthly (53 rows)daily (1.32k rows) Split (1)<br>train · 53 rows<br>train (53 rows)
SQL<br>Console
month<br>large_stringdate
2026-04-01 00:00:00<br>2026-06-01 00:00:00
agent<br>large_stringlengths
14
pct_requests<br>float64
67.8
pct_users<br>float64
54.3
2026-04
antigravity
3.0749
8.2763
2026-04
augment-cli
0.0215
0.0083
2026-04
claude-code
67.7732
54.0683
2026-04
cline
0.0966
0.3395
2026-04
codex
10.398
15.9169
2026-04
cowork
0.1507
0.1248
2026-04
cursor
1.5177
4.1806
2026-04
cursor-cli
9.0321
6.9233
2026-04
github-copilot
0.1795
0.1381
2026-04
openclaw
2.8046
3.0656
2026-04
opencode
0.0006
0.0017
2026-04
replit
0.1812
0.1315
2026-04
trae
0.2676
0.5941
2026-04
unknown
4.5017
6.2311
2026-05
antigravity
1.6912
6.7939
2026-05
augment-cli
0.0115
0.0074
2026-05
claude-code
6.405
5.7622
2026-05
cline
0.0569
0.3285
2026-05
codex
14.1461
20.8298
2026-05
cowork
0.0011
0.0025
2026-05
cursor
0.2773
0.9524
2026-05
cursor-cli
15.381
7.0018
2026-05
github-copilot
0.4042
0.2327
2026-05
goose
0.0001
0.0017
2026-05
openclaw
1.0793
1.6737
2026-05
opencode
0.2094
0.6124
2026-05
pi
0.3054
0.9895
2026-05
replit
0.0737
0.1238
2026-05
trae
0.1253
0.4069
2026-05
unknown
59.8326
54.2809
2026-06
antigravity
2.672
6.4335
2026-06
augment-cli
0.0015
0.0075
2026-06
claude-code
23.9004
25.3593
2026-06
cline
0.0837
0.2753
2026-06
codex
19.7824
18.4525
2026-06
cowork
0.0012
0.0023
2026-06
crush
0.0018
0.0121
2026-06
cursor
0.1613
0.1449
2026-06
cursor-cli
10.2491
6.3134
2026-06
gemini-cli
0.1257
0.172
2026-06
github-copilot
0.2054
0.2205
2026-06
goose
0.0006
2026-06
hermes-agent
1.3882
2.9939
2026-06
kilo-code
0.0412
0.0912
2026-06
kiro
0.1148
0.0537
2026-06
openclaw
2.3172
0.9102
2026-06
opencode
0.3981
0.8595
2026-06
pi
0.4466
0.9605
2026-06
replit
0.0796
0.0802
2026-06
trae
0.0942
0.3134
2026-06
unknown
37.6934
35.1833
2026-06
warp
0.1213
0.7463
2026-06
zed
0.1208
0.4139
Agent Usage on the Hugging Face Hub
Coding agents are real users of the Hugging Face Hub. Claude Code, Codex, Cursor, and a growing list of harnesses are searching for models, building and pushing datasets, training models on Jobs, spinning up Spaces — tens of millions of requests so far (hf CLI for agents). Now there's public data on which ones.
Requests made through the huggingface_hub library (including the hf CLI) carry an agent/ User-Agent token identifying the harness. This dataset publishes each harness's share of that agent-attributed traffic , month by month and day by day, updated by a scheduled HF Job.
Named harnesses ranked by share of requests, data through 2026-06 · updated 2026-07-02. The Dataset Viewer at the top of this page lets you browse, sort, and filter both tables — no code needed.
What you can see
Who's calling the Hub — the monthly leaderboard of named harnesses, and how it shifts as new tools launch and register.
Usage styles — compare request share with user share. An agent with 30% of requests but 8% of users is a small crowd running heavy automated pipelines; the reverse means many users, each doing a little.
Day-by-day detail — the daily config picks up what monthly numbers smooth over: launch spikes, growth curves, weekday-vs-weekend patterns.
Get your harness on the board
If you build a harness, register it to make sure your agent isn't missed — unregistered tools are counted only as unknown.
Attribution is automatic: huggingface_hub detects registered harnesses from environment variables and reports them in the User-Agent. To register, follow Register your agent harness — a Pull Request adding your tool to agent-harnesses.ts. No release is needed on either side: installed clients refresh the registry within a day, and your harness appears from the next monthly snapshot.
Only traffic through the Python huggingface_hub library (including the hf CLI) is attributed; direct HTTP calls to the Hub API are not counted. To confirm detection works, run inside your harness:
python -c "from huggingface_hub.utils import build_hf_headers; print(build_hf_headers()['user-agent'])"<br># should contain...