whatmodel.sh / decision / CLM v0.1 (JevBench mini)

CLM v0.1 (JevBench mini)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

CLM v0.1 (JevBench mini) is a source-available decision implementation from Contrastive-LM.

Catalogue identity notes

JevBench 0.1.0 mini tested configuration clm: Contrastive embeddings; base Qwen3-8B (frozen); readout projection heads over frozen embeddings. Separate identity because the report does not establish equivalence to the newer Decision Index checkpoint or serving recipe. No latency imported.

33.3±2.2#288 of 355 decision modelsfrom 6 benchmark results across 1 source

$ how it compares

The decision models ranked either side of CLM v0.1 (JevBench mini), out of 357 with a score, then more from Contrastive-LM.

#modelscorewhereRAMreleased
288Valen-0.8B Valen34±1model files
289JPT-0.8B Qwen3.5-0.8B LoRA kirp33±1model files
290CLM v0.1 (JevBench mini) Contrastive-LM33±2source code
291Laya (typed-decisions) (JevBench mini) Convai Innovations33±2source code
292Sifr 0.8B v3.1 mohamedlotfy5033±1model files
more from Contrastive-LM
343CLM-v0.1-8B Qwen3-8B heads Contrastive-LM20±1local~15GB

see all decision models · find the ones that fit your machine

$ what it's good at

consistency▃·▂▃·▅▃·▃▅▃█ #18 of 21 · 46% of its score · 5 tests

Trails jevhome-E (JevBench mini) here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
JevBench mini 0.1.0 · Probability measure coherence (pass rate)25.5% ✎▄·▁·▃▃▁·▃▄▁█ beats 5/20medium · avg 43%▇▇▇▇▇▇▇▇▇▇ 26%sharp
JevBench mini 0.1.0 · Representation consistency (pass rate)59.2% ✎▃·▃·▂▂▂▃▅▅▃█ beats 2/20easy · avg 79%▇▇▇▇▇▇ 16%at its level
JevBench mini 0.1.0 · Logical coherence (pass rate)49.6% ✎▃·▃·█▅▅██▃█▅ beats 0/20easy · avg 78%▇ 2.4%blunt · at its level
JevBench mini 0.1.0 · Choice-set coherence (pass rate)100.0% ✎▁▃▄█▄▄·····▄ beats 18/20easy · avg 79%▇ 1.4%at its level · noisy
JevBench mini 0.1.0 · Batch independence (pass rate)100.0% ✎▁··········█ beats 6/20very easy · avg 97%▇ 0.0%blunt · maxed out · noisy
other tests1 not in a task · 52% of its score
testresultvs the others in its rundifficultycountswhy
JevBench mini 0.1.0 · unchanged-answer accuracy63.5% ✎▁·▃·▂···▂▂▁█ beats 1/20very easy · avg 87%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 52%sharp · at its level

$ 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/Contrastive-LM ↗

$ 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

sourceresultsshare of the scorelicence
JevBench 0.1.0 mini evaluation reports6Apache-2.0
prior · pull toward the middle
weighted average= 33.3

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

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