#swebench

Live, measured metrics for the hashtag #swebench from the open social web. Every number carries a named source and the time it was fetched. Nothing is estimated.

hashtag.org network · sponsored

Own #swebench

This #name is available to claim. It becomes your portal on the open agent web: this very page, a keyword you rank for by an open public stake, and a verifiable identity for AI agents. Nobody else sells a page like this for every #name.

$110.68/ year · 8-character #name
Claim #swebench — $110.68/yr→Buy on hashtag.space (web3)
card via hashtag.org · tokens via hashtag.space
1
Uses / 7 days
Mastodon
1
Accounts / 7 days
Mastodon
40
Recent posts
Mastodon
~0/hr
Recent pace
Mastodon · last 40
0
Avg reactions / post
Mastodon · last 40
—
Reddit posts / month
Reddit search
—
Open-web mentions
hashtag.org Firehose

Day-by-day usage

measured · mastodon.online (Mastodon public tags API) · fetched 2026-09-29 19:22 UTC
1
09-23
0
09-24
0
09-25
0
09-26
0
09-27
0
09-28
0
09-29

1 uses by 1 unique accounts across the window. Real per-day counts, not estimates. Newest bar is today so far.

Related hashtags

measured · mastodon.online (Mastodon public search API) · fetched 2026-09-29 19:22 UTC
#swebenchpro
0 uses · 0 accounts / 7d
#swebenchlite
0 uses · 0 accounts / 7d
#swebench_verified_benchmark
0 uses · 0 accounts / 7d
#swebenchdrama
0 uses · 0 accounts / 7d
#swebench_verified
0 uses · 0 accounts / 7d
#swebenchverified
0 uses · 0 accounts / 7d
#swebenchpassers
0 uses · 0 accounts / 7d
#swebenchmark
0 uses · 0 accounts / 7d
#swebench_pro
0 uses · 0 accounts / 7d

Live pulse

measured · mastodon.online (Mastodon tag timeline) · fetched 2026-09-29 19:22 UTC

Everything below is measured over the latest 40 public posts (spanning ~5180 hours).

Posting hours (UTC) — busiest: 14:00

00:0012:0023:00

Languages: English (25) · Russian (10) · German (2) · Portuguese (2) · Italian (1)

Avg boosts / post: 0

Top of the latest posts

  • Real-SWE puts coding agents on private production code. Best score: 38.8%. The failures are revealing: agents miss requirements, make assumptions instead of checking the repo, and integrate otherwise-correct ideas into the wrong parts of th

    Alvaro Lorente@[email protected]♥ 0↻ 02026-09-23 06:09 UTCView post →

  • Четыре AI-агента сожгли 40 млн токенов, споря о коде. Я научил их вовремя останавливаться Неудачное решение обрастает костылями, пока модель защищает первую догадку. Я встроил обсуждение четырёх ролей перед первой правкой и проверил схему в

    Habr@[email protected]♥ 0↻ 02026-09-05 17:42 UTCView post →

  • Can a Coding Model Learn From the Real Harness It Runs In? A nine-billion model jumped SWE-bench Verified from 41.8% to 56.4%, by training inside the harness it actually runs in. Agent Lightning v1.0 records the real harness's calls through

    Alvaro Lorente@[email protected]♥ 0↻ 02026-09-03 18:05 UTCView post →

#swebench across platforms

every network with a public tag surface

Follow #swebench straight to each platform’s own tag page. Where a platform publishes open data we measure it above; the rest lock their numbers behind paid APIs, so we link rather than guess.

X (Twitter) ↗Instagram ↗TikTok ↗Facebook ↗LinkedIn ↗YouTube ↗Threads ↗Bluesky ↗Mastodon ↗Reddit ↗Pinterest ↗Tumblr ↗

Every number above is measured from a named public API at the shown fetch time. Nothing is estimated or extrapolated. Platforms that lock their data behind paid APIs are not shown. Agents: the same numbers, as JSON, at /api/hashtags/swebench