Opinion page / measured product / rank 5
Lovable Reviews and Ratings (2026)
Reader ratings for Lovable, aggregated on the same 0 to 100 scale we use for the lab index, with our own commentary underneath. If you are looking for Lovable opiniones, reviews or a rating you can check, the measurements behind every claim are on the Lovable benchmark page.
Reader index
82.79
6 voters, 55 axis scores. Never blended into the lab index.
Lab index
77.87
Measured on the vibeOps spec
Strict pass rate
82.2%
37 clean and 2 failed executions
Median per prompt
85.2 s
p90 133.1 s
Reader index
One score per reader per axis on a 0 to 100 scale, editable at any time.
Reader index
82.79
Lab index
77.87
6 readers, 55 axis scores, last change 14 Aug 2026. Readers rate Lovable 4.9 points above the measurement.
| Axis | Weight | Lab | Readers | Votes | Reader mean |
|---|---|---|---|---|---|
| Agent performance | 18% | 89.0 | 95.3 | 6 | |
| Reliability | 18% | 82.2 | 87.5 | 6 | |
| Scalability | 13% | 78.0 | 85.5 | 6 | |
| SEO and GEO | 12% | 75.0 | 78.0 | 6 | |
| API and MCP | 9% | 66.0 | 61.8 | 4 | |
| Integrations | 8% | 86.0 | 92.2 | 6 | |
| Design output | 7% | 92.0 | 99.6 | 5 | |
| Speed | 7% | 36.9 | 42.2 | 6 | |
| Value | 5% | 74.0 | 79.8 | 6 | |
| Code ownership | 3% | 79.0 | 85.3 | 4 |
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.
Score Lovable
Your scores feed the reader column only. Submitting an axis again edits your existing score rather than adding a second vote.
Reader scores need an account so one person cannot vote twice. Create a reader account or sign in.
Reader scores are published in a separate reader index column. They never move the measured lab index.
Reader lab notes on Lovable
Written by readers with an account. Sorted by upvotes, then by date.
Great to the demo, then I wrote the operations layer myself
Marijke Kuipers | 04 Aug 2026
Ran the whole vibeOps set against it for a client project rather than for the lab. The first three prompts were genuinely impressive: the tenancy constraint survived a refactor I did by hand afterwards, which is not something I expected. Where it cost me time was prompt five. The webhook it produced delivered fine and signed nothing, and when I asked for signing it added an HMAC header without a retry ladder, so a single 500 on the receiving end lost the event permanently. I ended up writing the delivery table, the retry schedule and the replay endpoint myself, which was two days. My scores reflect that split: high on agent performance and design, low on API and MCP.
Tobias Schneider | 05 Aug 2026
Matches what I saw. The signing header is there once you ask, but nothing stores a delivery attempt, so you cannot answer the question a customer actually asks, which is whether event 4821 was delivered.
Sergio de la Cruz | 06 Aug 2026
For the work I do none of that comes up, and the interface it produces saves me a week per project. Worth saying that the same product can be a nine or a five depending on which half of the spec you live in.
Write a lab note
What you built, what broke, what you measured. Short opinions belong in the open rating form further down.
Open rating distribution
Open ratings submitted without an account. Neither the lab index nor the reader index is part of this average.
| Band | Share | Ratings |
|---|---|---|
| 90 to 100 | 1 | |
| 80 to 89 | 1 | |
| 70 to 79 | 1 | |
| 60 to 69 | 0 | |
| 0 to 59 | 0 |
Lab commentary
Our reading of the numbers, kept separate from the reader ratings.
Strongest single shot agent in the cohort. It read the tenancy requirement in prompt 1 without being told twice and carried the organisation scope through every later prompt. It lost points on the machine facing surface: webhook signing needed an explicit follow up in 3 of 5 runs, and the generated API had no idempotency handling.
On the measured axes, Lovable is strongest at design output with a subscore of 92.0 and weakest at speed at 36.9. Because speed carries 7 of the 100 index points, that weakness costs it roughly 4.42 points against a perfect result on that axis alone.
Reader ratings and lab measurements answer different questions. A reader rating carries the thing a harness cannot capture: whether the product was pleasant to work with over weeks, how support behaved, whether the bill matched the plan. The lab index carries the thing an opinion cannot: 45 timed executions of the same specification with the failures counted. Read them side by side and treat a large gap between the two as the interesting signal.
3 open ratings
Newest first, no account needed. Moderated for spam and vendor astroturfing, not for sentiment.
Best design output we tested internally
Priya Raman, Product lead | 11 Aug 2026
91Our own bake off matched the lab result. The generated interface needed almost no visual clean up.
Credits go fast on long builds
Tom Reeve, Contract developer | 06 Aug 2026
79Quality is high but a full build burns the monthly allowance quickly. Budget for top ups.
Fast to a demo, slower to production
Marta Ibanez, Founder, 2 person SaaS | 02 Aug 2026
86Got a working product in an afternoon. The last 20 percent, webhooks and rate limits, took three days of manual work.
Add an open rating
No account needed. For a score that counts toward the reader index, use the axis form above.