whatmodel.sh / decision / jevhome-ettin-1b (JevBench mini)
jevhome-ettin-1b (JevBench mini)
↗ share
+ contribute
jevhome-ettin-1b (JevBench mini) is a source-available decision implementation from Jibril Frej.
Catalogue identity notes
JevBench 0.1.0 mini tested configuration jevhome-ettin-1b: Encoder, cross-encoder; base Ettin-encoder-1B; readout cross-encoder distilled from a Qwen3.5-4B teacher. Separate identity because the report does not establish equivalence to the newer Decision Index checkpoint or serving recipe. No latency imported.
$ how it compares
The decision models ranked either side of jevhome-ettin-1b (JevBench mini), out of 357 with a score, then more from Jibril Frej.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 254 |  vLLM-SR d3-lite (0.8B) | 42±1 | local | ||
| 255 |  OneJev1.5-0.8B | 42±4 | model files | ||
| 256 | jevhome-ettin-1b (JevBench mini) | 42±2 | source code | ||
| 257 | vLLM-SR vela3-fast (Vela 3.0, 0.4B) | 42±1 | model files | ||
| 258 | Cam1 | 42±1 | model files | ||
| more from Jibril Frej | |||||
| 293 | jevhome-E (JevBench mini) | 33±2 | source code | ||
see all decision models · find the ones that fit your machine
$ what it's good at
consistency▃·▂▂·▅▃·▃▅▃█ #15 of 21 · 56% of its score · 5 tests
| test | result | vs the others in its run | counts |
|---|---|---|---|
| JevBench mini 0.1.0 · Probability measure coherence (pass rate) | 29.2% ✎ | ▄·▁·▃▃▁·▃▄▁█ beats 6/20 | 38% |
| JevBench mini 0.1.0 · Representation consistency (pass rate) | 70.3% ✎ | ▃·▂·▂▂▃▃▅▅▃█ beats 5/20 | 15% |
| JevBench mini 0.1.0 · Logical coherence (pass rate) | 70.8% ✎ | ▃·▃·█▅▅██▃█▅ beats 5/20 | 2.3% |
| JevBench mini 0.1.0 · Choice-set coherence (pass rate) | 50.0% ✎ | ▁▃▄█▄▄·····▄ beats 2/20 | 1.3% |
| JevBench mini 0.1.0 · Batch independence (pass rate) | 100.0% ✎ | ▁··········█ beats 5/20 | 0.0% |
other tests1 not in a task · 42% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| JevBench mini 0.1.0 · unchanged-answer accuracy | 82.3% ✎ | ▁·▂·▂···▃▂▁█ beats 6/20 | 42% |
$ where to run it
Implementation and setup instructions ↗
Published implementation and setup documentation verified. This entry describes a decision readout or evaluated configuration, not necessarily separate trained weights. Follow the referenced version and model settings; hardware compatibility has not been verified. Source · checked 2026-10-31.
Model repository and documentation: github.com/Jibril-Frej ↗
$ how the score was worked out · JevBench 0.1.0 mini evaluatio… 98%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 JevBench 0.1.0 mini evaluation reports | 6 | 98% |
| prior · pull toward the middle | 2% | |
| weighted average | = 42.3 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json