Imajev-4B
Photos checked against your record, with a trained “can’t tell”.
4B
Two photos Can’t tell Typed questions
Imajev-4B against Jev mohit67890/imajev · also 2B and 9BTwo open image decision models · updated October 26, 2026
Imajev-4B vs Jev-Omni
Both read a photo, take your typed questions and return a probability for every option instead of prose. Each is ahead on a different leaderboard: Imajev-4B is far ahead of Jev-Omni on Image JevBench, and Jev-Omni ranks far higher on the text JevBench. Here is why, and which one fits your decisions.
Jev-Omni is hosted in our Playground in beta. Imajev-4B is not offered on this site, and neither model is available through our public API.
Jev-Omni
One classifier for text, images, audio and video.
12B
Images Audio Video
Explore Jev-Omni akhilaaa3/Jev-Omni · hosted hereTry it on this page
Run Jev 1.14 right here
Edit the example or paste your own text, then run it. Sign in to run: new accounts get 20 welcome credits, and a failed run costs nothing.
Jev-Omni has been offline since 19 October 2026, so this editor runs Jev 1.14, the model we recommend today. It also reads images.
Photo of an invoice · Jev 1.14
1 Input
A supplier sent this scan instead of a PDF: `scan`
2 Questions
- Yes / No Every line of the total is readable without guessing.
- Choice What kind of document is this?
- Score How badly does this need to be re-sent as a PDF?
- Box the total amount in `scan`
Measured by the same third party
Two leaderboards, opposite orders
Benchmark Heaven runs both boards. On images, Imajev-4B is near the top of the open-weights systems and well clear of Jev-Omni on both accuracy axes, while Jev-Omni still answers faster. On text, Jev-Omni ranks 81 places higher, #52 against #133 on the open-weights board, and scores 21 points higher on intelligence.
| Measure | Imajev-4B | Jev-Omni |
|---|---|---|
| Image JevBench v0.3.0 · images, 1,500 decisions | ||
| Rank and composite | #4 of 64 · 63.7 | #16 of 64 · 32.8 |
| Capability rank | #7 · 68.0 | #25 · 54.3 |
| Intelligence | 49.9 | 37.5 |
| Calibration | 86.1 | 71.0 |
| Adjusted median response | 0.37s | 0.32s |
| JevBench v1.7.48 · text, 1,500 decisions | ||
| Rank and composite | #133 of 235 · 30 | #52 of 235 · 71 |
| Capability rank | #103 · 61.9 | #52 · 71.3 |
| Intelligence | 34.6 | 55.5 |
| Calibration | 89.2 | 87.0 |
| Sealed half | 34.2 | 51.9 |
Version 0.3.0 is a fresh start, not an update: a new pool of 2,441 items on 2,160 images, drawn as 300 public and 1,200 sealed decisions, of which 713 items are real and 1,728 synthetic. 67 systems were measured and 64 ranked; two more keep a separately dated v0.1.5 score and are not ranked on the v0.3 scale, and one is an unranked second read of H2O-Lightning-4B v1.4.1 under changed serving settings. No v0.3.0 score can be compared with a v0.1.5 one, and the per-system public and sealed item counts the old release published are not part of this one. Wity-1, the closed API that led v0.1.5, was re-measured on the new pool on 7 October at 71.2, and on 16 October Sqwish’s closed kettei-fat API entered above it at 75.4. With those two ahead, Imajev-4B then held the highest place of any system whose weights you can download. The same day kettei-chub, Sage 1.4 with images and Celeris-1 Decision joined too, all hosted APIs, and on 17 October the open-weights basal-omni-1.0-ultra, a text model reading images zero-shot, entered at #11, followed by Drex Max on 21 October at #8. On 27 October H2O.ai’s open-weights H2O-Lightning-4B v1.4.1 (image) entered at #3, above Imajev-4B. Together these additions pushed Jev-Omni down to #16. A 15 October addendum added TypeSafe’s two Jev-with-images preview rows, removed one row at its operator’s request and moved the board’s Jev-class cost reference to Jev with images, which widened the pool behind its Capability ranking and moved the Capability places above. No composite changed.
Read the test, not just the number
The Imajev author’s own image test
ImajevBench is the Imajev author’s benchmark: 279 test questions about photos, records and text, 21 of which have no answer the evidence supports. It is a preview with AI-generated images and no human audit yet.
83.9% vs 78.5%
Imajev-4B against Jev-Omni, run by the Imajev author with Jev-Omni’s own predict(). The author’s significance test, run on an earlier Imajev release, found a gap of this size not significant.
18 vs 0 of 21
The gap comes from the “can’t tell” questions, which Jev-Omni has no output for. On answerable questions the author found Jev-Omni slightly ahead of an earlier Imajev release, and better calibrated.
What actually differs
Both return decision probabilities instead of generated explanations. Their inputs, limits and serving paths are not the same.
| Capability | Imajev-4B | Jev-Omni |
|---|---|---|
| Backbone | Qwen3.5-4B with a LoRA adapter and decision readout; also 2B and 9B | Gemma 4 12B IT, fine-tuned and merged |
| Images | Up to two per request, such as a reference and a target | One per request |
| Audio and video | Not supported | Video as 16 sampled frames; the model also takes audio, switched off on Jev AI after our tests |
| Can’t tell | A trained unknown probability and an abstained flag on every answer | No abstain output |
| Request format | TypeSafe-style /v1/systemone, plus images | Its own Python loader and predict(); hosted here in the Playground |
| Options | 2 to 254 per question, 1 to 8 questions per request | Head accepts 256; best supported at 20 or fewer |
| Hardware | Apple silicon through MLX or one CUDA GPU; 9.3 GB base weights | A CUDA GPU; about 24 GB of BF16 weights |
| Licence | Apache-2.0 adapters and code on an Apache-2.0 base | Apache-2.0 weights following Gemma 4; dataset rights separate |
| Try it | The author’s Hugging Face demo, or run it yourself | Hosted in this site’s Playground (sign in), or run it yourself |
Which should you evaluate?
Start with Imajev-4B when…
- You compare a photo with a record: a listing, a return, a part against a known-good one.
- You want the model to say it cannot tell, and send those cases to a person.
- Most of your decisions are about images, where it places far above Jev-Omni on the image board.
- You need to run it locally, including on a Mac.
Start with Jev-Omni when…
- Your decisions involve video, not only photos. For audio, use Clef-Omni.
- Many of your decisions are text-only, where it scores far higher.
- Response time on image decisions matters most.
- You want to try it right now without a GPU, in this site’s Playground.
Imajev-4B vs Jev-Omni FAQ
Which is better on images, Imajev-4B or Jev-Omni?
Imajev-4B, and by a wider margin than before. On Image JevBench v0.3.0 it is fourth of 64 at 63.7, behind only kettei-fat at 75.4, Wity-1 at 71.2 and H2O-Lightning-4B v1.4.1 (image) at 66.6, against Jev-Omni’s 32.8 at #16, with intelligence 49.9 against 37.5 and calibration 86.1 against 71.0. On that board’s Capability ranking it is seventh at 68.0, with Jev-Omni #25 at 54.3. Jev-Omni answers faster. On the Imajev author’s own ImajevBench, Imajev-4B scores 83.9% against 78.5%.
Which is better on text?
Jev-Omni, by a wide margin. On the text JevBench v1.7.48 it is #52 of 235 open-weights systems at 71, with intelligence 55.5, and #52 on the headline Capability ranking; Imajev-4B is #133 at 30, with intelligence 34.6 and #103 on Capability.
Can either model say it cannot tell?
Imajev-4B can: every answer carries a trained unknown probability and an abstained flag. Jev-Omni has no abstain output, so on questions whose honest answer is “can’t tell” it always picks one of the options.
Can I try them without a GPU?
Jev-Omni is hosted in this site’s Playground: sign in and run text, image or video decisions; audio runs on Clef-Omni. Imajev-4B has a demo on Hugging Face Spaces run by its author, and it also runs locally on Apple silicon.
Are these made by TypeSafe?
No. Imajev is a third-party project by Mohit Garg and Jev-Omni a third-party model by akhilaaa3. Both borrow Jev’s typed-decision idea; TypeSafe’s Jev itself reads text only.
Sources read October 26, 2026: JevBench v1.7.48, Image JevBench v0.3.0, Imajev repository and Jev-Omni model card. Imajev and Jev-Omni belong to their respective authors and are not affiliated with Jev AI or TypeSafe.