whatmodel.sh / decision / deepseek-v4.1-flash (DecisionBench openrout)

deepseek-v4.1-flash (DecisionBench openrout)

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

deepseek-v4.1-flash (DecisionBench openrout) is an open-weights decision model from DeepSeek. Its published size is 552B parameters. Licence: mit.

Catalogue identity notes

DecisionBench v1 comparison: deepseek/deepseek-v4.1-flash, revision openrouter-service-snapshot-2026-09-21-reasoning-aware-v1; language-model; adapter openrouter-chat-completions-json-schema-reasoning-aware-v1; probability source model-emitted-complete-candidate-probability-vector. 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.

58.0±1.7#130 of 355 decision modelsfrom 27 benchmark results across 1 source

$ how it compares

The decision models ranked either side of deepseek-v4.1-flash (DecisionBench openrout), out of 357 with a score.

#modelscorewhereRAMreleased
130AJev lora7 (Gemma 4 12B) andyzhang23258±1model files
131Winnow-12B (DecisionBench b6ac22b0) EldanRing58±2local~16GB
132deepseek-v4.1-flash (DecisionBench openrout) DeepSeek58±2model files
133Nace.AI Drex v1.5 Nace.AI58±1model files
134Clef-flash Cloudflare58±1api $0.038/M input tokens +local~25GB2026-10

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

$ what it's good at

workflows▁▁▁▂▁▂▂▆█▄▁▁ #35 of 114 · 99% of its score · 27 tests

Beats vLLM-SR d3-flash (9B), Clef-omni, Jebadiah 27B v2.1 (Ollaya Q4_K_M) here, though they rank higher overall. Trails StartLux-Decision v1.1 2B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
DecisionBench v1 · action selection (all-request accuracy) submitted59.4% ✎▂▁▂▃▄▃▁▁▂█▇▄ beats 40/56medium · avg 51%▇▇▇▇▇▇▇▇ 21%at its level
DecisionBench v1 · routing (all-request accuracy) submitted75.0% ✎▁▁▁▁▂▂▁▁▁▁▃█ beats 41/56medium · avg 64%▇▇▇▇▇ 13%sharp · at its level
DecisionBench v1 · navigation (all-request accuracy) submitted71.0% ✎▁▁·▃▁▁▁▁▁▁▃█ beats 39/56medium · avg 58%▇▇▇▇ 11%sharp · at its level
DecisionBench v1 · record classification (all-request accuracy) submitted42.0% ✎▃▂▁▂▂▂▄▁▃▆██ beats 30/56medium · avg 39%▇▇▇▇ 10%at its level
DecisionBench v1 · verification (all-request accuracy) submitted62.0% ✎▁▁▅▁▃▂▂▁▁▇▇█ beats 44/56medium · avg 56%▇▇▇ 6.5%blunt · at its level
DecisionBench v1 · document record classification (all-request accuracy) submitted48.4% ✎▂▂·▁▂▄▃▆▄█▁▁ beats 37/56medium · avg 46%▇▇ 5.4%at its level
DecisionBench v1 · rubric scoring prioritization (all-request accuracy) submitted46.1% ✎▁▁▅▅▃▃▅█▃▃·▁ beats 33/56medium · avg 45%▇▇ 4.5%at its level
DecisionBench v1 · triage (all-request accuracy) submitted88.0% ✎▁·▁·▂▃▄▃▂█▄▂ beats 49/56easy · avg 78%▇▇ 4.0%sharp
19 more tests
testresultvs the others in its rundifficultycountswhy
DecisionBench v1 · completion assessment (all-request accuracy) submitted93.0% ✎▁▁▂▃▄·▂▁▁▆█▂ beats 46/56very easy · avg 86%▇ 3.6%sharp · maxed out
DecisionBench v1 · content classification (all-request accuracy) submitted79.0% ✎▁····▁▁▁▁▁▁█ beats 37/56easy · avg 73%▇ 3.1%typical
DecisionBench v1 · review scoring (all-request accuracy) submitted54.0% ✎▁······▂▃▅▇█ beats 32/56medium · avg 52%▇ 3.0%at its level
DecisionBench v1 · bounded extraction (all-request accuracy) submitted93.4% ✎▃▃▂·▁▁▁▁▁▁█▅ beats 33/56very easy · avg 86%▇ 2.9%sharp · maxed out · noisy
DecisionBench v1 · retrieval ranking (all-request accuracy) submitted55.0% ✎▁▁·▁▁▂▃▂▃▁▁█ beats 38/56medium · avg 35%▇ 2.4%at its level · noisy
DecisionBench v1 · target selection (all-request accuracy) submitted48.0% ✎▁▁·▁▁▁▃▁▂▁▁█ beats 50/56hard · avg 29%▇ 2.3%at its level · noisy
DecisionBench v1 · risk scoring (all-request accuracy) submitted56.0% ✎▁▁·▁▂▂▁▃▃█▇▃ beats 21/56medium · avg 62%▇ 2.0%blunt · at its level
DecisionBench v1 · reasoning (all-request accuracy) submitted76.9% ✎▂·▁▁▆█▅▃▃▃▁▁ beats 53/56medium · avg 52%▇ 1.2%at its level
DecisionBench v1 · document workflows (all-request accuracy) submitted68.2% ✎█▄▃▁▃▃▂▂▁·▁▁ beats 37/56easy · avg 66%▇ 0.8%blunt · at its level
DecisionBench v1 · retrieval verification (all-request accuracy) submitted87.9% ✎█▁▂▁▁▁▁▁▂▅▁▁ beats 50/56easy · avg 74%▇ 0.7%noisy
DecisionBench v1 · guardrails moderation (all-request accuracy) submitted63.1% ✎▁▁▁▂▆▃▃█···▁ beats 46/56medium · avg 52%▇ 0.5%blunt · at its level · noisy
DecisionBench v1 · safety gating (all-request accuracy) submitted95.0% ✎▂·····▆█▂▂▄▇ beats 43/56easy · avg 77%▇ 0.4%noisy
DecisionBench v1 · document classification (all-request accuracy) submitted34.0% ✎▂▁·▂▁▁▁▄▁▂▂█ beats 45/56hard · avg 21%▇ 0.4%blunt · noisy
DecisionBench v1 · semantic filtering (all-request accuracy) submitted86.0% ✎▁·▁▁·▁▁█▂▃█▂ beats 50/56easy · avg 71%▇ 0.3%blunt · noisy
DecisionBench v1 · content moderation (all-request accuracy) submitted99.0% ✎▁····▃▅▃▂▂·█ beats 41/56easy · avg 82%▇ 0.2%noisy
DecisionBench v1 · context management (all-request accuracy) submitted88.5% ✎▁·····▁▇▆▁█▃ beats 40/56easy · avg 73%▇ 0.2%blunt · noisy
DecisionBench v1 · entity alignment (all-request accuracy) submitted86.7% ✎▁·▁▄▇▄▅▄▃▁▄█ beats 35/56easy · avg 66%▇ 0.2%at its level · noisy
DecisionBench v1 · function agent skill routing (all-request accuracy) submitted86.2% ✎▃▃▄▅▆▂▅▄▃▄█▃ beats 33/56easy · avg 68%▇ 0.1%blunt · at its level · noisy
DecisionBench v1 · routing triage (all-request accuracy) submitted59.2% ✎█▁···▁·▂▆▅▂▂ beats 36/56medium · avg 42%▇ 0.0%blunt · at its level · noisy

$ where to run it

Download model files ↗
Published model files verified. This is a verified repository snapshot; the benchmark does not establish an identical weight revision. 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/deepseek-ai ↗

$ 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= 58.0

$ 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