Asynchronous cloud coding agent that clones repositories into Google-hosted VMs, plans changes, tests work, and prepares patches.
Attributed output
GitHub public-attribution count, 2026-06-09
1.0MAttributed outputGitHub public-attribution count, 2026-06-09Commits and PRs on public GitHub provably produced by this agent. Exact if it is visible through its own bot or workflow, an estimate if measured through attribution it declares (these can swing 30 to 40 percent between measurement runs at multi-million scale), and a floor either way.Index rank · all agents
#4 of 40 CLI agents
#4Index rank · all agents#4 of 40 CLI agentsRanked by agent-opened PRs in the trailing 4 complete weeks, matching the index. Agents without a tracked PR series follow, ordered by total attributed output. Other filters or sorts change the list position.n/a
Business signal
No sourced business metric
n/aBusiness signalNo sourced business metric$0 limited
Entry price
Daily and concurrent task quotas
$0 limitedEntry priceDaily and concurrent task quotasNot published
GitHub stars
No public repo
Not publishedGitHub starsNo public repoNot published
npm demand/wk
Public npm counter
Not publishednpm demand/wkPublic npm counterWeekly package downloads as an install-demand floor; adjusted when the vendor's main install channel is no longer npm.From the public record
488 PRs · 281 repositories
Search window: – · captured Oct 14, 2026 (UTC)
40.4%197 / 488 sampled PRs
197 of 488 PR titles use the labels Palette, Bolt or Sentinel, across 117 repositories. Examples name accessibility, performance and security tasks. The labels are observable; they do not establish whether a human or a schedule triggered the work.
Separate study: CodeRabbit follow-ups, May 25–October 11, 2026
In CodeRabbit’s May 25–October 11 audit, we found 1 original PRs crediting Jules, followed by 2 CodeRabbit auto-fix PRs; 1 follow-ups were merged. This cohort was selected through CodeRabbit, so it cannot tell us what fraction of this agent’s overall work receives follow-up fixes. Credits can name a model without identifying its client.
BintzGavin/helios: “✨ RENDERER: Inline `limit` evaluation in `CaptureLoop.ts` runWorker paths” (merged). rwilliamspbg-ops/Ghostlink: “🎨 Palette: Enhance keyboard focus rings and aria-current selection on Projects tab” (merged).
This is a sample of publicly attributed work, not total usage or a market-share estimate. Repositories are not customers. Attribution is declared in metadata, not independently verified tool execution; it does not prove exclusive authorship or autonomy. PR status is recorded at capture; merged does not establish correctness.
Up to 100 most recently created search results in each of five consecutive 28-day windows; post-filtered for attribution in the PR body, bot author, or Codex branch. Not a random or exhaustive sample.
Returned records dated Jun 21, 2026–Oct 11, 2026.
Download the dated audit evidence (JSON)Weekly attributed output on public GitHub. Output visible through the agent's own bot or workflow is exact; output measured through declared attribution is an estimate. Window: 2025-02-03 to 2026-10-12.
Public GitHub commits attributed to Jules
Jules's work is visible through its workflow, so these are exact.
Last full week (ending 2026-10-18): 13,666 signed commits (-2.8% week over week), exact count.
Distinct users: 1,397 humans opened the 12,600 most recently sampled PRs, about 9 PRs per person.
Merge rate: 63.4% of its pull requests get merged (126,147 of 199,070 over 8 settled weeks, 2026-08-10 to 2026-10-04). The window ends about ten days back so recently-opened PRs have had time to merge.
GitHub review activity: 3.3% of its merged PRs had at least one GitHub review record before merging; the other 96.7% had none. This is a broad weekly signal and can include bot reviews; the human-review read below comes from the enriched PR sample.
In significant repos (company-owned or 10+ stars), 12.5% of its merged PRs got a human review. Across all sampled repos, the rate is 22.5%. Human review means a non-author, non-bot GitHub reviewer in the PR-lifecycle sample.
Median time to merge: 57 min. The unreviewed median is 1.89h. The human-reviewed median is 0.59h. From a 120-PR sample merged 2026-07-30 to 2026-08-28.
Test-inclusion: 25% of its sampled merged PRs touch a test file. This is a rough read on production readiness.
Commits per PR: 0.6 over the last 8 weeks. Both counts come from the same visible workflow, so this is a clean measure of how many commits make up one delegated task.
Flow share: 0.3% of the 1999 most recent public PRs sampled (5 PRs, 2 windows, last 2026-09-27).
Task lens: 1.2M agent-opened PRs. Jules changed how it opens pull requests in late January 2026; the current measure matches the earlier one exactly, so the series is consistent across all time. 51% of these PRs are merged.
Snapshot 2026-06-09, confidence High, tracked since 2026-06-09, +1% over 7 days. Read the methodology.
Share of Jules's own classified sampled work by segment; countries reflect where its PR authors are based. Full breakdowns: Languages, Frameworks, Industries, Map.
Pricing snapshot: 2026-06-07. Check the vendor for current terms.
Jules is task-quota priced through Google AI plan access rather than token-priced per task.
No token budget per Jules task is published; limits are expressed as rolling daily tasks and concurrent task count.
| Plan | Price | Allowance | Unit | Token / Overage Notes |
|---|---|---|---|---|
| Jules | $0 | 15 daily tasks; 3 concurrent tasks. | tasks | |
| Jules in Google AI Pro | $19.99/mo Google AI Pro | 100 daily tasks; 15 concurrent tasks. | tasks | |
| Jules in Google AI Ultra | $249.99/mo Google AI Ultra | 300 daily tasks; 60 concurrent tasks. | tasks |
Summary checked 2026-07-03.
Jules is tied to Google AI plan access with Free, Pro, and Ultra-style task limits described in Jules usage docs.
Jules is a Google product, funded through Alphabet (NASDAQ: GOOGL).
Compliance follows Google Cloud's certifications. Consumer access ships through Google AI plans; enterprise terms apply for paid workspaces.