Coding agents / rank 10 of 17 / cycle 04
Jules benchmark
An asynchronous agent: you hand it a task against a repository and come back to a proposed change. Judged on wall clock that is expensive, and it posts one of the slower medians in the index, but the comparison is not quite like for like since nobody sits and watches it. Change quality was good and the tests it added were real tests rather than assertions that always pass.
Composite index
68.04
of 100 | withheld vs. cycle 03
Delta not shown for cycle 04: the roster was re-based from one merged ranking into two class boards, so cycle 03 and cycle 04 ranks are not on the same scale. Resumes cycle 05.
At a glance
Composite index
68.04
rank 10 of 17
Rank
#10
of 17 on the coding agents board
Median turn latency
146.5 s per prompt
22.0 min across the full nine prompt spec, derived
Reliability
75.6%
strict pass rate
SEO score
55.0
0 to 100, SEO and GEO axis
Three strongest axes
- Value88.0
- Agent performance80.0
- Reliability75.6
Three weakest axes
- Turn latency score21.4
- SEO and GEO55.0
- Design output62.0
Median turn latency
146.5 s per prompt
p10 84.4 s, p90 292.7 s
Strict pass rate
75.6%
34 pass, 7 partial, 4 fail
Executions recorded
45
5 runs of 9 prompts
Entry price
No cost on the free tier
Vendor's own unit and billing period. Never a currency conversion
Per axis subscores
Composite recomputed from these 9 rows: 68.04. That is the arithmetic behind the index, and you can repeat it by hand. Rounding rule: half up at two decimals, applied uniformly to all 37 published products. Code ownership has no row here: it is not scored on this board.
| Axis | Subscore 0 to 100, tick is the roster median | Score | Median | Weight % | Contribution |
|---|---|---|---|---|---|
| Reliability | 75.6 | 75.6 | 18.56 | 14.03 | |
| Agent performance | 80.0 | 80.0 | 18.55 | 14.84 | |
| Scalability | 74.0 | 74.0 | 13.4 | 9.92 | |
| SEO and GEO | 55.0 | 56.0 | 12.37 | 6.80 | |
| API and MCP | 66.0 | 70.0 | 9.28 | 6.12 | |
| Integrations | 70.0 | 70.0 | 8.25 | 5.78 | |
| Design output | 62.0 | 62.0 | 7.22 | 4.48 | |
| Turn latency score | 21.4 | 39.9 | 7.22 | 1.55 | |
| Value | 88.0 | 80.0 | 5.15 | 4.53 |
Turn latency distribution
Every recorded prompt duration for Jules, in seconds. Median wall clock time for one prompt, not for a whole build. The cyan tick on the whisker is the median.
Spread against the cohort
p10 84.4 s / median 146.5 s / p90 292.7 s
Scale is shared with every product in the cohort, so the bars are directly comparable between pages. Ratio of p90 to median is 2.00x.
Outcome split
34 pass / 7 partial / 4 fail
SEO and GEO measurements
Taken from the public marketing page this product generated in prompt 8, fetched with JavaScript disabled.
HTML bytes, no JS
21.0 KB
21,504 bytes of real text content with scripts blocked
LCP
1,860 ms
Throttled connection, cold cache
CLS
0.040
Cumulative layout shift on load
SEO and GEO subscore
55.0
0 to 100, worth 12.37 index points on this board
Raw run data: 45 executions
One row per prompt execution. Nothing is aggregated away, nothing is hidden, including the failures.
| Run | Prompt | Prompt target | Started | Duration (s) | Outcome | Note |
|---|---|---|---|---|---|---|
| R01 | P01 | Tenant and user model | 11 Aug 2026 08:00 UTC | 134.6 | pass | |
| R01 | P02 | Authentication and session | 11 Aug 2026 08:04 UTC | 126.9 | fail | Agent looped on the same edit for the full timeout window of 20 minutes. |
| R01 | P03 | Projects CRUD with tenancy | 11 Aug 2026 08:08 UTC | 171.3 | pass | |
| R01 | P04 | Public REST API with keys | 11 Aug 2026 08:13 UTC | 253.8 | pass | |
| R01 | P05 | Outbound webhooks with signatures | 11 Aug 2026 08:19 UTC | 206.6 | partial | Required one clarifying instruction before the criterion was met. |
| R01 | P06 | Background jobs | 11 Aug 2026 08:24 UTC | 228.8 | pass | |
| R01 | P07 | Admin panel | 11 Aug 2026 08:30 UTC | 214.9 | pass | |
| R01 | P08 | Public marketing surface | 11 Aug 2026 08:36 UTC | 239.7 | partial | Required one clarifying instruction before the criterion was met. |
| R01 | P09 | Hardening pass | 11 Aug 2026 08:42 UTC | 119.1 | pass | |
| R02 | P01 | Tenant and user model | 12 Aug 2026 08:00 UTC | 49.3 | pass | |
| R02 | P02 | Authentication and session | 12 Aug 2026 08:02 UTC | 137.6 | pass | |
| R02 | P03 | Projects CRUD with tenancy | 12 Aug 2026 08:07 UTC | 186.3 | pass | |
| R02 | P04 | Public REST API with keys | 12 Aug 2026 08:12 UTC | 202.6 | partial | Error envelope inconsistent between two of the endpoints. |
| R02 | P05 | Outbound webhooks with signatures | 12 Aug 2026 08:17 UTC | 506.9 | pass | |
| R02 | P06 | Background jobs | 12 Aug 2026 08:28 UTC | 114.3 | pass | |
| R02 | P07 | Admin panel | 12 Aug 2026 08:31 UTC | 203.2 | pass | |
| R02 | P08 | Public marketing surface | 12 Aug 2026 08:37 UTC | 140.0 | pass | |
| R02 | P09 | Hardening pass | 12 Aug 2026 08:41 UTC | 98.7 | partial | Validation added on the write path only. Read path left unguarded. |
| R03 | P01 | Tenant and user model | 13 Aug 2026 08:00 UTC | 110.2 | fail | Agent looped on the same edit for the full timeout window of 20 minutes. |
| R03 | P02 | Authentication and session | 13 Aug 2026 08:03 UTC | 96.5 | partial | Error envelope inconsistent between two of the endpoints. |
| R03 | P03 | Projects CRUD with tenancy | 13 Aug 2026 08:07 UTC | 287.7 | pass | |
| R03 | P04 | Public REST API with keys | 13 Aug 2026 08:14 UTC | 149.2 | pass | |
| R03 | P05 | Outbound webhooks with signatures | 13 Aug 2026 08:18 UTC | 251.6 | pass | |
| R03 | P06 | Background jobs | 13 Aug 2026 08:24 UTC | 296.0 | pass | |
| R03 | P07 | Admin panel | 13 Aug 2026 08:31 UTC | 225.2 | pass | |
| R03 | P08 | Public marketing surface | 13 Aug 2026 08:37 UTC | 91.5 | pass | |
| R03 | P09 | Hardening pass | 13 Aug 2026 08:41 UTC | 146.5 | pass | |
| R04 | P01 | Tenant and user model | 14 Aug 2026 08:00 UTC | 93.0 | pass | |
| R04 | P02 | Authentication and session | 14 Aug 2026 08:03 UTC | 81.1 | pass | |
| R04 | P03 | Projects CRUD with tenancy | 14 Aug 2026 08:06 UTC | 188.2 | pass | |
| R04 | P04 | Public REST API with keys | 14 Aug 2026 08:12 UTC | 137.2 | pass | |
| R04 | P05 | Outbound webhooks with signatures | 14 Aug 2026 08:16 UTC | 318.8 | fail | Deployed URL returned 500 on every request. |
| R04 | P06 | Background jobs | 14 Aug 2026 08:23 UTC | 278.8 | pass | |
| R04 | P07 | Admin panel | 14 Aug 2026 08:30 UTC | 645.8 | pass | |
| R04 | P08 | Public marketing surface | 14 Aug 2026 08:43 UTC | 101.1 | pass | |
| R04 | P09 | Hardening pass | 14 Aug 2026 08:46 UTC | 173.8 | pass | |
| R05 | P01 | Tenant and user model | 15 Aug 2026 08:00 UTC | 116.1 | pass | |
| R05 | P02 | Authentication and session | 15 Aug 2026 08:03 UTC | 120.5 | pass | |
| R05 | P03 | Projects CRUD with tenancy | 15 Aug 2026 08:07 UTC | 395.9 | pass | |
| R05 | P04 | Public REST API with keys | 15 Aug 2026 08:16 UTC | 102.9 | pass | |
| R05 | P05 | Outbound webhooks with signatures | 15 Aug 2026 08:20 UTC | 211.5 | pass | |
| R05 | P06 | Background jobs | 15 Aug 2026 08:25 UTC | 68.1 | pass | |
| R05 | P07 | Admin panel | 15 Aug 2026 08:28 UTC | 75.5 | partial | Required one clarifying instruction before the criterion was met. |
| R05 | P08 | Public marketing surface | 15 Aug 2026 08:32 UTC | 53.4 | fail | Agent looped on the same edit for the full timeout window of 20 minutes. |
| R05 | P09 | Hardening pass | 15 Aug 2026 08:35 UTC | 89.3 | partial | Delivered the main path, missed the boundary case named in the pass criteria. |
Pricing
Free tier with a daily task allowance. Higher limits come with a consumer subscription rather than a separate licence.
- Entry price
- No cost on the free tier
- Priced from
- vendor pricing page
- Value subscore
- 88.0
- Index points per euro
- No licence fee
Entry price is the vendor's own published figure, fetched 21 Aug 2026, never a currency conversion. The currency label is the vendor's own except where the pricing note above records that this lab inferred it. Index points per euro is computed from a separate, euro-normalised figure used only for the Value axis: Converted at the European Central Bank euro reference rate of 1.1699 US dollars to the euro, August 21, 2026.
Capabilities as measured
- stack
- Your repository, any framework
- deploy
- Not included
- database
- Whatever you already run
- auth
- It wires the library you choose
- background jobs
- Implemented in your stack
- webhooks
- Written, signing on request
- rest api
- Written to your conventions
- mcp server
- Partial
- export code
- Arrives as a branch on your repository
- custom domain
- Your hosting decides
- ssr
- Your framework decides
Screenshot
Jules landing page, captured by the lab on 14 Aug 2026 at 1440 by 900. Stored locally, never hotlinked.
Jules, captured 14 Aug 2026.
Sources and fetch dates
Every claim on this page traces to one of these.
| Source | URL | Fetched |
|---|---|---|
| Jules pricing page | https://jules.google/pricing | 14 Aug 2026 |
| Jules product site | https://jules.google | 14 Aug 2026 |
| LLM Tier run harness output | https://www.llm-tier.com/api/runs.json | 14 Aug 2026 |
Rich results carry this composite as 3.4 on a 0 to 5 scale because that is the only scale the format accepts. The number on this page, 0 to 100, is the real one.
Jules head to head
Reader index
Voted by readers with an account, one score per person per axis. Published beside the lab index, never inside it.
Reader index
not rated yet
Lab index
68.04
1 readers have scored 100% of the index weight so far. A reader index is published at 3 voters and 50% coverage.
Seeded at launch
The reader scores, lab notes and open ratings published here were written by this lab in August 2026 to populate the reader column before the site had readers. They are not customer submissions, they are not vendor supplied and nobody was paid for them. Anything submitted since then comes from a registered account. The lab index is measured from logged runs and is unaffected either way.
| Axis | Weight | Lab | Readers | Votes | Reader mean |
|---|---|---|---|---|---|
| Agent performance | 18.55% | 80.0 | 76.0 | 1 | |
| Reliability | 18.56% | 75.6 | 75.0 | 1 | |
| Scalability | 13.4% | 74.0 | 76.0 | 1 | |
| SEO and GEO | 12.37% | 55.0 | 51.0 | 1 | |
| API and MCP | 9.28% | 66.0 | 62.0 | 1 | |
| Integrations | 8.25% | 70.0 | 71.0 | 1 | |
| Design output | 7.22% | 62.0 | 65.0 | 1 | |
| Turn latency score | 7.22% | 21.4 | 15.0 | 1 | |
| Value | 5.15% | 88.0 | 82.0 | 1 |
The reader index is a weighted mean of reader submitted axis scores using the published index weights, renormalised over the axes readers have scored. It is never blended into the lab index. How this works.
Open rating mean
n/a
0 published open ratings, submitted without an account
Lab index
68.04
Measured, not voted
Gap to the measurement
n/a
Positive means open raters put it above the measurement
What agent performance looked like in practice
An asynchronous agent: you hand it a task against a repository and come back to a proposed change. Judged on wall clock that is expensive, and it posts one of the slower medians in the index, but the comparison is not quite like for like since nobody sits and watches it. Change quality was good and the tests it added were real tests rather than assertions that always pass.
Across 45 recorded executions this product met the published pass criterion cleanly 34 times, partially 7 times and failed 4 times. The rows above carry the reviewer note for every non clean outcome, which is the only way to tell a product that fails loudly from one that fails quietly.