whatmodel.sh · find the best decision models you can run

Best Decision models for document & agent workflows for a 64GB graphics card

A leaderboard of the best Decision models for document & agent workflows for a 64GB graphics card (with 32GB of system RAM). The best one that runs on it right now is Perplexity Decider Flash v1.2 (4B), followed by Winnow-12B (DecisionBench b6ac22b0) and vLLM-SR d3-flash (9B). Published typed-decision evaluations, with evaluator-specific protocols and tested settings retained. Each task score uses only its reviewed benchmarks. Models that run on it show the download that fits and the command to run it.

43 models · 108 benchmarks · updated

Compare tested decision models and inference recipes across text, vision and workflow tasks. Use task filters and raw results: probability consistency measures coherence, not correctness.

Filter by your hardware and budget → ~64GB fast · reset
ram[2][4][8][16][24][32][48][64][96][128][192][256]
[]
GB
gpu[apple silicon][graphics card][unified memory][none]
vram[4][8][12][16][24][32][48][64][80][96][128]
[]
GB
price[local][$0.05][$0.1][any]max per M input tokens
Filter by task
task[all tasks][tools & automation][retrieval & classification][language understanding][arts & human taste][knowledge & reasoning][vision][workflows][consistency]

Scores use reviewed workflows evaluations only. Compare scores within this task. Models without task measurements are excluded.

New #1s for your hardware:
orRSS

detecting your OS…

$ terminal

detecting your OS…

Run curl "whatmodel.sh/decision?ram=32&gpu=card&vram=64&max=any"to get this list in your terminal (add &format=json for agents).

Score vs release date

Release dates available for 17 of 43 models. Undated models are omitted; month-only dates are plotted at the start of the month.

#modeltask score↓where↓RAM↓speed↓get itsources
#modeltask score↓where↓RAM↓speed↓get itsources
1Perplexity Decider v1.1 (27B) Perplexity80±1api price unknownopenrouter.ai ↗DecisionPivot · StartLux LabsLeanMCP · JevBench
3Mercury Decide (paid endpoint) Inception72±3api price unknownopenrouter.ai ↗LeanMCP · JevBenchPerch benchmark results
8Microsoft-Decision-1 Microsoft66±2api price unknownopenrouter.ai ↗LeanMCP · JevBenchSystem One Bench (Ali Alhashimi)Perch benchmark results
9Jev 1.13 TypeSafe65±1api $0.042/M input tokensdocs.typesafe.ai ↗Hanno Labs DecisionBench v1 reviewed resultsDecisionPivot · StartLux LabsLeanMCP · JevBenchTyped Decision Models under Prevalence Shift: Decision Training, Calibration, and Implicit PriorsSystem One Bench (Ali Alhashimi)DecisionModelHub · support benchmarks+3
12GPT-6 Luna Decisions OpenAI64±2api price unknownopenrouter.ai ↗LeanMCP · JevBenchSystem One Bench (Ali Alhashimi)DecisionModelHub · support benchmarksDecideBench (Cho Yin Yong)Perch benchmark results
15Jev 1.14 TypeSafe63±2api $0.042/M input tokensdocs.typesafe.ai ↗JevalKit (Valen Team)System One Bench (Ali Alhashimi)
17d1-ultra Liquid AI63±4*api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)
18d1 (hosted) Liquid AI62±2api price unknownopenrouter.ai ↗LeanMCP · JevBenchSystem One Bench (Ali Alhashimi)Perch benchmark results
24Clef Cloudflare61±1api $0.24/M input tokens +local~68GBdevelopers.cloudflare.com ↗DecisionPivot · StartLux LabsLeanMCP · JevBenchSystem One Bench (Ali Alhashimi)DecisionModelHub · support benchmarksDecideBench (Cho Yin Yong)
27Perplexity Decider Flash v1.2 (4B) · ollaya-f32 Perplexity61±2local~10GBhow to run › Ollaya · original runsSystem One Bench (Ali Alhashimi)
31Span-02 (Gemma 4 26B-A4B) Respan60±4*api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)
33d1.1 (hosted) Liquid AI60±4*api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)
36Mercury Decide (free endpoint) Inception59±4*api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)
38Solar Decide Upstage59±3api price unknownopenrouter.ai ↗LeanMCP · JevBenchSystem One Bench (Ali Alhashimi)
39Winnow-12B (DecisionBench b6ac22b0) · ollaya-q8-0 EldanRing58±1local~16GBhow to run › Hanno Labs DecisionBench v1 reviewed resultsOllaya · original runs
41vLLM-SR d3-flash (9B) · ollaya-f32 vLLM Semantic Router58±2local~18GBhow to run › Ollaya · original runs
45vLLM-SR d3-mini (4B) · ollaya-f32 vLLM Semantic Router57±2local~11GBhow to run › Ollaya · original runs
46Jebadiah 27B v2.1 (Ollaya Q4_K_M) · q4_k_m-ollaya frontier-infra57±4local~16GBhow to run › Ollaya · original runs
48Jebadiah 9B v2.1 (Ollaya Q8_0) · q8_0-ollaya frontier-infra57±4local~9.1GBhow to run › Ollaya · original runs
57Decider 4B · ollaya-f32 Mapika55±2local~11GBhow to run › Ollaya · original runs
59vLLM-SR Decision 2.0 Nox 4B · ollaya-f32 vLLM Semantic Router54±2local~12GBhow to run › Ollaya · original runs
61vLLM-SR Decision 2.0 Lux 9B · ollaya-f32 vLLM Semantic Router54±2local~20GBhow to run › Ollaya · original runs
65Jebadiah 4B v2 (Ollaya Q8_0) · q8_0-ollaya frontier-infra54±2local~5.5GBhow to run › Ollaya · original runs
66Kev 4B r10 · ollaya-f32 Jared Palmer54±1local~11GBhow to run › Ollaya · original runsJevalKit (Valen Team)System One Bench (Ali Alhashimi)
68Clef-flash · bf16 Cloudflare53±1local~25GBhow to run › DecisionPivot · StartLux LabsLeanMCP · JevBenchOllaya · original runsJevalKit (Valen Team)DecisionModelHub · support benchmarksSystem One Bench (Ali Alhashimi)+1
69Bespoke Nimble 9B v3 · ollaya-f32 Bespoke Labs52±2local~21GBhow to run › Ollaya · original runs
70Solar Decide Flash Upstage52±2api price unknownopenrouter.ai ↗LeanMCP · JevBenchSystem One Bench (Ali Alhashimi)
71Bespoke Nimble 9B v2 · ollaya-f32 Bespoke Labs52±2local~20GBhow to run › Ollaya · original runs
72Span-01 Pro Respan52±9api price unknownwww.respan.ai ↗Perch benchmark results
73Clef-omni Cloudflare52±2api $0.15/M input tokens~87GBdevelopers.cloudflare.com ↗LeanMCP · JevBenchSystem One Bench (Ali Alhashimi)
74baselabs joey-1 (2B) baselabs52±4*api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)
75vLLM-SR d3-nano (2B) · ollaya-f32 vLLM Semantic Router52±2local~8.2GBhow to run › Ollaya · original runsSystem One Bench (Ali Alhashimi)
79GLiClass v4 edge (0.4B) · ollaya-f32 Knowledgator Engineering49±2local~2.7GBhow to run › Ollaya · original runs
82vLLM-SR Decision 2.0 Sol 2B · ollaya-f32 vLLM Semantic Router46±2local~11GBhow to run › Ollaya · original runs
83vLLM-SR d3-lite (0.8B) · ollaya-f32 vLLM Semantic Router45±2local~4.5GBhow to run › Ollaya · original runsSystem One Bench (Ali Alhashimi)
84LiquidAI d1-omni-600M · ollaya-f32 Liquid AI45±2local~2.7GBhow to run › Ollaya · original runsSystem One Bench (Ali Alhashimi)
86Decision-1.0-Eos-0.8B (DecisionBench 3c2d6326) · ollaya-f32 llm-semantic-router44±1local~5.8GBhow to run › Hanno Labs DecisionBench v1 reviewed resultsOllaya · original runs
87vLLM-SR Decision 2.0 Eos 0.8B · ollaya-f32 vLLM Semantic Router44±2local~5.8GBhow to run › Ollaya · original runs
88vLLM-SR Decision 2.0 Kai 0.6B · ollaya-f32 vLLM Semantic Router43±2local~3.7GBhow to run › Ollaya · original runs
90Kev 0.8B r15 · ollaya-f32 Jared Palmer41±2local~4.6GBhow to run › Ollaya · original runsSystem One Bench (Ali Alhashimi)
92vLLM-SR d3-edge (0.6B) · ollaya-f32 vLLM Semantic Router40±2local~2.6GBhow to run › Ollaya · original runsSystem One Bench (Ali Alhashimi)
104laya (DecisionBench 1c5edc17) · ollaya-f16 ConvAI Innovations27±1local~1.7GBhow to run › Hanno Labs DecisionBench v1 reviewed resultsOllaya · original runs
113CLM-v0.1-8B Qwen3-8B heads · ollaya-f32 Contrastive-LM15±2local~15GBhow to run › Ollaya · original runsDecideBench (Cho Yin Yong)Perch benchmark results

+ 74 more don't fit your machine or budget · show them

117 of 117 catalogue models listed here · sources and licences

* new or thinly tested: ranked on too few benchmark results · † no independently run results verified · ~RAM is estimated; full context may need more memory

43 models · 108 benchmarks · updated