whatmodel.sh / decision / decider-2b v10 (JevBench mini)
decider-2b v10 (JevBench mini)
↗ share
+ contribute
decider-2b v10 (JevBench mini) is a source-available decision implementation from Mapika.
Catalogue identity notes
JevBench 0.1.0 mini tested configuration decider-2b: Decoder, answer-letter logits; base Qwen3.5-2B-Base; readout fine-tune + calibration-aware RL; option-letter logits. 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 decider-2b v10 (JevBench mini), out of 357 with a score, then more from Mapika.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 230 | Lev1 3.7B | 47±1 | model files | ||
| 231 | Ryotide Qwen v0.3.1 (decision readout) | 47±1 | source code | ||
| 232 | decider-2b v10 (JevBench mini) | 47±2 | source code | ||
| 233 | Jev-Style 2B v4 | 47±1 | model files | ||
| 234 |  Saracura PT-BR 4B | 46±1 | model files | ||
| more from Mapika | |||||
| 43 | Decider 31B (NVFP4) | 66±1 | model files | ||
| 85 | Decider chat · Gemma-4-31B (decision readout) | 62±1 | source code | ||
| 110 | decider-35b-a3b (DecisionBench d9783c58) | 60±2 | model files | ||
see all decision models · find the ones that fit your machine
$ what it's good at
consistency▃·▂▂·▅▃·▃▅▃█ #11 of 21 · 62% of its score · 5 tests
| test | result | vs the others in its run | counts |
|---|---|---|---|
| JevBench mini 0.1.0 · Probability measure coherence (pass rate) | 38.0% ✎ | ▄·▁·▃▃▁·▃▄▁█ beats 9/20 | 45% |
| JevBench mini 0.1.0 · Representation consistency (pass rate) | 78.6% ✎ | ▃·▂·▂▂▂▃▅▅▃█ beats 12/20 | 14% |
| JevBench mini 0.1.0 · Logical coherence (pass rate) | 67.1% ✎ | ▃·▃·█▅▅██▃█▅ beats 2/20 | 2.3% |
| JevBench mini 0.1.0 · Choice-set coherence (pass rate) | 75.0% ✎ | ▁▃▄█▄▄·····▄ beats 9/20 | 1.2% |
| JevBench mini 0.1.0 · Batch independence (pass rate) | 100.0% ✎ | ▁··········█ beats 8/20 | 0.0% |
other tests1 not in a task · 36% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| JevBench mini 0.1.0 · unchanged-answer accuracy | 84.9% ✎ | ▁·▂·▂···▂▃▁█ beats 8/20 | 36% |
$ 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/Mapika ↗
$ 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 | = 47.1 |
$ 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