Jev-Omni (akhilaaa3, Gemma-4-12B merged)
Jev rebuild · by akhilaaa3
JevBench v1.4.2.2 score
51.3
Rank #11 of 91 ranked systems.
12.0 points behind Jev 1.13.0's 63.3.
Published axes
- intelligence
- 46.8
- calibration
- 64.1
- speed
- 81.5
- cost
- 53.0
Against Jev 1.13.0
Four radars compare this fixed pair across the score axes, accuracy per tier, and accuracy by family on the hard tier and sealed set. Further out is better on every spoke.
- A: Jev-Omni — Jev rebuild · Score 51.3 (#11)
- B: Jev 1.13.0 — Jev (TypeSafe, closed) · Score 63.3 (#4)
The four score axes
Accuracy per tier, incl. sealed
Current question set by family (hard + sealed)
Sealed set by family
Availability and evidence
- Openness
- Code and weights marked open in the published row
- License note
- Apache-2.0, following Gemma 4; dataset rights stated separately by the author
- Cost evidence
- estimated; the board’s row disclosure contains the published basis.
- Endpoint condition
- our RunPod GPU (L40 48 GB, Czechia), reached over the internet from Germany
- Note on this row
- Akhilaaa3/Jev-Omni revision c050d51354147985d13286cf4acf90f562f2c631, the author's own load_model.py (merged text decision model + 256-way head) and his own predict(), transformers 5.17.0 / torch 2.8.0 from the pod image; built on the CPU and moved to CUDA with every nn.Linear weight cast to bfloat16 first - the same cast his reference loader jev_omni.py applies - because our 46 GB GPU cannot hold his fp32 copy; on our RunPod L40 in Czechia Offline local open-weight inference on the frozen 308-item v1.4 set in a network-disabled, read-only container; no operator endpoint received sealed text.
- Published source
- https://huggingface.co/akhilaaa3/Jev-Omni
From the public v1.4.2.2 aggregate. Scores and ranks can change when a new release is published.
Read the full board and published method. The overall score is a composite, not raw accuracy.