whatmodel.sh / decision / gpt-5.6-luna (DecisionBench openrout)

gpt-5.6-luna (DecisionBench openrout)

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

gpt-5.6-luna (DecisionBench openrout) is a closed-weights decision model from OpenAI, released 9 July 2026.

Catalogue identity notes

DecisionBench v1 comparison: openai/gpt-5.6-luna, 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.

Model information sources

Release date ↗

59.2±1.7comparison only: no current run option verifiedfrom 27 benchmark results across 1 source

$ how it compares

The decision models ranked either side of gpt-5.6-luna (DecisionBench openrout), out of 357 with a score, then more from OpenAI.

#modelscorewhereRAMreleased
117JADE theunnecessarythings59±1model files2026-10-01
118xor (DecisionBench 679decd4) juspay59±2model files
119gpt-5.6-luna (DecisionBench openrout) OpenAI59±2comparison only2026-07-09
120simplejev-qwen3.8-27b (DecisionBench 1d4bf0f2) featherless-ai59±2model files
121Clef-omni Cloudflare59±1api $0.15/M input tokens +local~87GB2026-10
more from OpenAI
94GPT-6 Luna Decisions OpenAI62±1api price unknown2026-10-06

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

$ what it's good at

workflows▁▁▁▂▁▂▂▆█▃▁▁ #34 of 116 · 99% of its score · 27 tests

Beats Mercury Decide (free endpoint), vLLM-SR d3-flash (9B) 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) submitted60.4% ✎▂▁▂▃▄▃▁▁▂█▇▄ beats 42/56medium · avg 51%▇▇▇▇▇▇▇▇ 21%at its level
DecisionBench v1 · routing (all-request accuracy) submitted76.3% ✎▁▁▁▁▂▂▁▁▁▁▃█ beats 45/56medium · avg 64%▇▇▇▇▇ 13%sharp
DecisionBench v1 · navigation (all-request accuracy) submitted74.0% ✎▁▁·▃▁▁▁▁▁▁▃█ beats 51/56medium · avg 58%▇▇▇▇ 11%sharp · at its level
DecisionBench v1 · record classification (all-request accuracy) submitted43.0% ✎▃▂▁▂▂▂▄▁▃▆██ beats 33/56medium · avg 39%▇▇▇▇ 10%at its level
DecisionBench v1 · verification (all-request accuracy) submitted63.0% ✎▁▁▅▁▃▂▂▁▁▇▇█ beats 48/56medium · avg 56%▇▇▇ 6.6%blunt · at its level
DecisionBench v1 · document record classification (all-request accuracy) submitted46.6% ✎▂▂·▁▂▄▃▆▄█▁▁ beats 35/56medium · avg 46%▇▇ 5.4%at its level
DecisionBench v1 · rubric scoring prioritization (all-request accuracy) submitted44.9% ✎▁▁▅▅▃▃▅█▃▃·▁ beats 30/56medium · avg 45%▇▇ 4.6%at its level
DecisionBench v1 · triage (all-request accuracy) submitted89.0% ✎▁·▁·▂▃▄▃▂█▄▂ beats 51/56easy · avg 78%▇▇ 3.9%sharp
19 more tests
testresultvs the others in its rundifficultycountswhy
DecisionBench v1 · completion assessment (all-request accuracy) submitted93.0% ✎▁▁▂▃▄·▂▁▁▆█▂ beats 43/56very easy · avg 86%▇ 3.4%sharp · maxed out
DecisionBench v1 · content classification (all-request accuracy) submitted80.0% ✎▁····▁▁▁▁▁▁█ beats 43/56easy · avg 73%▇ 3.1%typical
DecisionBench v1 · review scoring (all-request accuracy) submitted61.0% ✎▁······▂▃▅▇█ beats 44/56medium · avg 52%▇ 3.0%at its level
DecisionBench v1 · bounded extraction (all-request accuracy) submitted92.9% ✎▃▃▂·▁▁▁▁▁▁█▅ beats 29/56very easy · avg 86%▇ 2.7%sharp · maxed out · noisy
DecisionBench v1 · retrieval ranking (all-request accuracy) submitted60.0% ✎▁▁·▁▁▂▃▂▃▁▁█ beats 51/56medium · avg 35%▇ 2.5%at its level · noisy
DecisionBench v1 · target selection (all-request accuracy) submitted49.0% ✎▁▁·▁▁▁▃▁▂▁▁█ beats 52/56hard · avg 29%▇ 2.4%at its level · noisy
DecisionBench v1 · risk scoring (all-request accuracy) submitted52.0% ✎▁▁·▁▂▂▁▃▃█▇▃ beats 15/56medium · avg 62%▇ 2.0%blunt · at its level
DecisionBench v1 · reasoning (all-request accuracy) submitted86.2% ✎▂·▁▁▆█▅▃▃▂▁▃ beats 55/56medium · avg 52%▇ 1.3%at its level
DecisionBench v1 · document workflows (all-request accuracy) submitted73.4% ✎█▄▃▁▁▃▂▂▁·▁▁ beats 41/56easy · avg 66%▇ 0.8%blunt · at its level
DecisionBench v1 · retrieval verification (all-request accuracy) submitted87.2% ✎█▁▂▁▁▁▁▁▂▅▁▁ beats 43/56easy · avg 74%▇ 0.7%noisy
DecisionBench v1 · guardrails moderation (all-request accuracy) submitted61.8% ✎▁▁▁▂▆▃▃█···▁ beats 40/56medium · avg 52%▇ 0.6%blunt · at its level · noisy
DecisionBench v1 · safety gating (all-request accuracy) submitted97.0% ✎▂·····▆█▂▂▄▇ beats 49/56easy · avg 77%▇ 0.4%noisy
DecisionBench v1 · document classification (all-request accuracy) submitted38.0% ✎▂▁·▂▁▁▁▄▁▂▂█ beats 53/56hard · avg 21%▇ 0.4%blunt · noisy
DecisionBench v1 · semantic filtering (all-request accuracy) submitted83.0% ✎▁·▁▁·▁▁█▂▃█▂ beats 39/56easy · avg 71%▇ 0.3%blunt · noisy
DecisionBench v1 · content moderation (all-request accuracy) submitted99.0% ✎▁····▃▅▃▂▂·█ beats 39/56easy · avg 82%▇ 0.2%noisy
DecisionBench v1 · context management (all-request accuracy) submitted90.5% ✎▁·····▁▇▆▁█▃ beats 44/56easy · avg 73%▇ 0.2%blunt · noisy
DecisionBench v1 · entity alignment (all-request accuracy) submitted81.1% ✎▁·▁▄▇▄▅▄▂▁▄█ beats 33/56easy · avg 66%▇ 0.2%at its level · noisy
DecisionBench v1 · function agent skill routing (all-request accuracy) submitted81.1% ✎▃▃▄▅▆▂▅▄▃▄█▃ beats 31/56easy · avg 68%▇ 0.1%blunt · at its level · noisy
DecisionBench v1 · routing triage (all-request accuracy) submitted51.8% ✎█▁···▁·▂▆▅▂▂ beats 30/56medium · avg 42%▇ 0.0%blunt · at its level · noisy

$ where to run it

No verified API or download recorded yet · add one

Model repository and documentation: openrouter.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= 59.2

$ 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