🔍 Read the full analysis: A 30-Image Test Drive Of Jev’s Visual AI on ThorstenMeyerAI.com
Get business pricing on office and shipping supplies
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Thorsten Meyer tested Jev 1.14 and 1.13 on 30 featured images after TypeSafe added opt-in image input and a bounding-box output type. In the reported product-image results, Jev 1.14 matched photos to their own headlines and identified nine of 10 products; the source account is cut off before it gives results for the 15 news images and five essays.
Thorsten Meyer tested TypeSafe’s Jev 1.14 on 30 featured images from his website after the company added opt-in image input and a bounding-box response type. His published results show strong separation between matching product photos and unrelated headlines, but the account is cut off before reporting the outcomes for the 15 news images and five essays in the test.
TypeSafe released Jev 1.14 on October 14, 2026, adding image support behind an explicit request header. The company also introduced a new box type that returns one bounding box and a confidence score for a described target. TypeSafe says image input also works with Jev 1.13, describing that earlier model as “multimodal all along.”
Meyer tested 30 recent images from ThorstenMeyerAI.com: 10 product roundups, 15 news posts and five essays. For each image, he asked whether it fit its own headline, a randomly selected other headline, and an unrelated headline about repotting a monstera plant. The 10 product images also received a multiple-choice question asking which product each showed. He ran the test on both model versions, using about 60 requests in total.
For the product images, Meyer reports that Jev 1.14 gave each image’s own headline a score from 0.65 to 0.89, with a median of 0.76. Random other headlines scored no higher than 0.18, with a median of 0.06; the plant headline scored no higher than 0.10. Jev 1.14 picked the correct product in nine of 10 cases. Jev 1.13 got all 10 product identifications right, though two had confidence below 0.5.
Jev 1.14 · images + box
Jev can see now. Here is what 30 of my images showed.
I asked about every featured image three times: does it fit its own headline, a random other post’s headline, an off-topic headline? Both Jev versions, about 60 requests.
Does this image fit the headline?
Median “yes” probability on Jev 1.14, by kind of post
- Product roundups10 images0.760.061.13: 0.79 / 0.05
- News posts15 images0.330.161.13: 0.51 / 0.25
- My essays5 images0.510.161.13: 0.62 / 0.33
Off-topic headline (“How to Repot a Monstera Plant”): never above 0.10 on either version, 60 of 60 rejected.
Why news is weak: almost all news images show the same scene, people at desks looking at monitors. Jev could not tell them apart, and it was right not to.
1.14 is stricter than 1.13
Same 30 images, same questions
| Result | 1.14 | 1.13 |
|---|---|---|
| Fits own headline (≥ 0.5) | 15/30 | 23/30 |
| Fits a wrong headline (≥ 0.5) | 1/30 | 4/30 |
| Own beats random other (AUC) | 0.91 | 0.89 |
| Names the product (10 options) | 9/10 | 10/10 |
1.14’s one product miss: mesh system read as “Wi-Fi 7 router”, confidence 0.60. Re-tune thresholds before jev-latest moves.
What shipped on Oct 14
- headerImagesin state, instructions or criteria · 1.13 + 1.14 ·
TypeSafe-Danger-Zone-Images - 1.14 onlyBox typeone bounding box + confidence. 6 of 6 clear targets boxed (0.66–0.90); an absent bicycle still got a box (0.03)
- headerLong context64k / 32k → 262k tokens, slower and less accurate
- opt-inHelp find weaknessesqueries used for testing, not training
now: jev-preview = 1.14~Oct 21: jev-latest → 1.14notice: 60 days (1.13), 30 days (1.14)
Source: own test on October 15, 2026. 30 featured images from ThorstenMeyerAI.com (10 product roundups, 15 news posts, 5 essays), JPEG at up to 1,568 px, jev-1.14.0 and jev-1.13.0 with the image header. Bars show medians. AUC = probability that an image’s own headline scores higher than a random other headline. Features, prices and limits: TypeSafe blog post “Jev 1.14” and docs.typesafe.ai, as of October 15, 2026.
Powered by Thorsten Meyer AI
What the Product Results Show
The test offers an early, small-scale indication of how a structured decision model handles images, rather than a broad benchmark of visual AI. In Meyer’s product sample, Jev 1.14 separated relevant from irrelevant headlines and mostly identified the pictured product. That kind of repeatable classification could be useful for tasks such as checking whether an image fits a page title.
The results also show why confidence scores and human review matter. Jev 1.14’s one product-identification error was a plausible mix-up: the image showed a router with antennas alongside a cylindrical mesh unit, and the model chose “Wi-Fi 7 router” with 0.60 confidence. Meyer says every correct answer had confidence of at least 0.77, suggesting a threshold might have flagged the miss in this sample. That is an observation from 10 images, not evidence that the same threshold will work reliably in other datasets.
As an affiliate, we earn on qualifying purchases.
Jev Adds Opt-In Image Input
Jev is designed to answer typed questions about supplied information, including yes-or-no decisions with probabilities, choices from a list and scores on a scale. Before this release, those questions were text-based. The image feature lets an image reference appear in the request’s state, a question’s instructions or the criteria used for choices and scoring.
TypeSafe has placed the new capabilities behind what it calls a “danger zone” header, saying the features are not ready for production use. The company said there is “still way too much jaggedness” to vouch for them in production. Image input uses the header TypeSafe-Danger-Zone-Images: true; the box output is available only on Jev 1.14.
TypeSafe also added an opt-in longer context setting, raising request and question limits from 64,000 and 32,000 tokens to 262,000 each. It cautions that reliability can fall on long inputs. The release is available as jev-preview; TypeSafe said jev-latest would move to it about a week after release, around October 21.
““There is still way too much jaggedness for us to vouch for these for production use cases.””
— TypeSafe, in its release post, as quoted by Thorsten Meyer
As an affiliate, we earn on qualifying purchases.
News Image Results Are Missing
The source account ends as it begins to discuss the news-image results, with the line “The news posts were a dif”. It provides no completed findings for those 15 images or for the five essays, so the overall 30-image test cannot be evaluated from the available material. The results also do not establish performance beyond this website’s image selection and prompts.
TypeSafe has not vouched for the new image capabilities in production. The source does not provide independent benchmarks, a larger validation set, or details about how consistent the scores would be across repeated runs. Meyer’s results are a reported test, not a guarantee of performance on other images or tasks.
As an affiliate, we earn on qualifying purchases.
Preview Rollout and Further Testing
TypeSafe said jev-preview was available following the October 14 release and that jev-latest was scheduled to move to Jev 1.14 about one week later, around October 21. It also described deprecation notice periods of 60 days for Jev 1.13 and 30 days for Jev 1.14, while saying a formal policy was still to come.
For readers weighing the test’s relevance, the next useful evidence would include the missing news and essay results, broader evaluations, and TypeSafe’s later guidance on production readiness. The company advises users to run their own evaluations; the source does not give a date for a production endorsement or for the formal deprecation policy.
AI-powered image bounding box tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What did Jev add in version 1.14?
TypeSafe added opt-in image input and a new box response type that returns one bounding box and a confidence score. Image input is also supported on Jev 1.13, according to the company.
What images were included in the test?
Meyer tested 30 featured images from his site: 10 product roundups, 15 news posts and five essays. The published source gives detailed results for the product group only.
How did Jev 1.14 perform on product images?
In Meyer’s reported sample, Jev 1.14 matched each product image to its own headline with scores from 0.65 to 0.89 and identified the correct product in nine of 10 cases. These are results from a small, site-specific test.
Is TypeSafe recommending the image features for production?
No. TypeSafe said the features still have too much “jaggedness” for it to vouch for production use, and requires an explicit opt-in header.
Source: ThorstenMeyerAI.com
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
