whatmodel.sh / decision / KaLM-Reranker-V1-Small-R2 (DecisionBench d1c8fdf9)

KaLM-Reranker-V1-Small-R2 (DecisionBench d1c8fdf9)

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

KaLM-Reranker-V1-Small-R2 (DecisionBench d1c8fdf9) is an open-weights decision model from KaLM-Embedding. Its published size is 2.115977456B parameters. Licence: apache-2.0.

Catalogue identity notes

DecisionBench v1 comparison: KaLM-Embedding/KaLM-Reranker-V1-Small-R2, revision d1c8fdf974e904da12f700e700eca6c868426001; classifier; adapter jev-compatible-server-v1; probability source native_probability_distribution. Kept separate from newer or unpinned Decision Index recipes; standard prompt, not compact. No unverified runtime or OpenRouter price imported. Current repository metadata checked separately; an unpinned alias is not evidence that today’s weights reproduce this historical run.

Model information sources

· Weights licence ↗ · Parameter count ↗

Published served parameter count; not a RAM or download-size measurement.

17.6±1.8#349 of 355 decision modelsfrom 27 benchmark results across 1 source

$ how it compares

The decision models ranked either side of KaLM-Reranker-V1-Small-R2 (DecisionBench d1c8fdf9), out of 357 with a score.

#modelscorewhereRAMreleased
349system-one-gemma akash-kamat18±1source code
350Supersonic Labs Julia 1 Supersonic Labs18±1model files2026-09-25
351KaLM-Reranker-V1-Small-R2 (DecisionBench d1c8fdf9) KaLM-Embedding18±2model files
352KaLM-Reranker-V1-Nano-R2 (DecisionBench 3902d645) KaLM-Embedding17±2model files
353Nirnaya 0.4B kshitijthakkar16±1model files2026-10-06

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

$ what it's good at

workflows▁▃▁▂▁▂▂▆█▄▁▁ #109 of 114 · 99% of its score · 27 tests

Beats CLM-v0.1-8B Qwen3-8B heads, Supersonic Labs Julia 1 here, though they rank higher overall.

testresultvs the others in its rundifficultycountswhy
DecisionBench v1 · action selection (all-request accuracy) submitted25.0% ✎▂▁▂▃▄▃▁▁▂█▇▄ beats 10/56medium · avg 51%▇▇▇▇▇▇ 15%typical
DecisionBench v1 · completion assessment (all-request accuracy) submitted45.0% ✎▁▁▂▃▄·▂▁▁▆█▂ beats 10/56very easy · avg 86%▇▇▇▇▇ 12%sharp · at its level
DecisionBench v1 · routing (all-request accuracy) submitted13.7% ✎▁▃▁▁▂▂▁▁▁▁▃█ beats 3/56medium · avg 64%▇▇▇▇▇ 12%sharp
DecisionBench v1 · bounded extraction (all-request accuracy) submitted23.3% ✎▃▃▂·▁▁▁▁▁▁█▅ beats 5/56very easy · avg 86%▇▇▇ 8.1%sharp · noisy
DecisionBench v1 · triage (all-request accuracy) submitted47.0% ✎▁·▁·▂▃▄▃▂█▄▂ beats 11/56easy · avg 78%▇▇▇ 7.7%sharp · at its level
DecisionBench v1 · verification (all-request accuracy) submitted35.0% ✎▁▁▅▁▃▂▂▁▁▇▇█ beats 6/56medium · avg 56%▇▇▇ 6.8%blunt · at its level
DecisionBench v1 · navigation (all-request accuracy) submitted5.0% ✎▃▁·▃▁▁▁▁▁▁▃█ beats 0/56medium · avg 58%▇▇▇ 6.7%sharp
DecisionBench v1 · record classification (all-request accuracy) submitted15.0% ✎▃▃▁▂▂▂▄▁▃▆██ beats 5/56medium · avg 39%▇▇ 5.9%typical
19 more tests
testresultvs the others in its rundifficultycountswhy
DecisionBench v1 · content classification (all-request accuracy) submitted23.0% ✎▁····▁▁▃▁▁▁█ beats 5/56easy · avg 73%▇▇ 4.6%at its level
DecisionBench v1 · document record classification (all-request accuracy) submitted16.3% ✎▂▃·▁▂▄▃▆▄█▁▁ beats 5/56medium · avg 46%▇▇ 4.0%typical
DecisionBench v1 · rubric scoring prioritization (all-request accuracy) submitted15.0% ✎▃▁▅▅▃▃▅█▃▃·▁ beats 0/56medium · avg 45%▇ 3.4%typical
DecisionBench v1 · review scoring (all-request accuracy) submitted27.0% ✎▁······▂▃▅▇█ beats 6/56medium · avg 52%▇ 2.6%at its level
DecisionBench v1 · risk scoring (all-request accuracy) submitted41.0% ✎▁▁·▁▂▃▁▃▃█▇▃ beats 8/56medium · avg 62%▇ 2.3%blunt · at its level
DecisionBench v1 · retrieval verification (all-request accuracy) submitted49.1% ✎█▁▂▁▁▁▁▁▂▅▁▁ beats 6/56easy · avg 74%▇ 1.2%at its level · noisy
DecisionBench v1 · reasoning (all-request accuracy) submitted10.8% ✎▃·▁▁▆█▅▃▃▂▁▁ beats 2/56medium · avg 52%▇ 1.1%at its level
DecisionBench v1 · document workflows (all-request accuracy) submitted46.7% ✎█▄▃▁▁▃▂▂▁·▁▁ beats 1/56easy · avg 66%▇ 1.0%blunt · at its level
DecisionBench v1 · safety gating (all-request accuracy) submitted52.0% ✎▂·····▆█▂▂▄▇ beats 14/56easy · avg 77%▇ 0.8%at its level · noisy
DecisionBench v1 · retrieval ranking (all-request accuracy) submitted1.0% ✎▁▃·▁▁▂▃▂▃▁▁█ beats 3/56medium · avg 35%▇ 0.8%too hard for it · noisy
DecisionBench v1 · target selection (all-request accuracy) submitted2.0% ✎▁▁·▃▁▁▃▁▂▁▁█ beats 3/56hard · avg 29%▇ 0.7%too hard for it · noisy
DecisionBench v1 · guardrails moderation (all-request accuracy) submitted11.7% ✎▃▁▁▂▆▃▃█···▁ beats 0/56medium · avg 52%▇ 0.5%blunt · at its level · noisy
DecisionBench v1 · semantic filtering (all-request accuracy) submitted59.0% ✎▁·▁▁·▁▁█▂▃█▂ beats 22/56easy · avg 71%▇ 0.4%blunt · at its level · noisy
DecisionBench v1 · content moderation (all-request accuracy) submitted66.0% ✎▁····▃▅▃▂▂·█ beats 19/56easy · avg 82%▇ 0.4%at its level · noisy
DecisionBench v1 · context management (all-request accuracy) submitted72.0% ✎▁·····▁▇▆▁█▃ beats 30/56easy · avg 73%▇ 0.3%blunt · at its level · noisy
DecisionBench v1 · document classification (all-request accuracy) submitted18.0% ✎▂▁·▂▁▁▁▄▁▃▂█ beats 27/56hard · avg 21%▇ 0.2%blunt · noisy
DecisionBench v1 · entity alignment (all-request accuracy) submitted100.0% ✎▁·▁▄▇▄▅▄▂▁▄█ beats 52/56easy · avg 66%▇ 0.2%at its level · noisy
DecisionBench v1 · function agent skill routing (all-request accuracy) submitted95.5% ✎▃▃▄▅▆▂▅▄▃▄█▃ beats 42/56easy · avg 68%▇ 0.2%blunt · at its level · noisy
DecisionBench v1 · routing triage (all-request accuracy) submitted63.4% ✎█▁···▁·▂▆▅▂▂ beats 39/56medium · avg 42%▇ 0.0%blunt · at its level · noisy

$ where to run it

Download model files ↗
Published model files verified. This is the revision recorded by the benchmark. A local run command and hardware requirements have not been verified for this entry. Source · checked 2026-10-31.

Model repository and documentation: huggingface.co/KaLM-Embedding ↗

$ how the score was worked out · Hanno Labs DecisionBench v1 r… 99%

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
Hanno Labs DecisionBench v1 reviewed results · submitted; independent execution unverified27CC0 1.0
prior · pull toward the middle
weighted average= 17.6

$ 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