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

Best Decision models for images & screens for a 32GB graphics card

A leaderboard of the best Decision models for images & screens for a 32GB graphics card (with 32GB of system RAM). The best one that runs on it right now is Clef-flash, followed by Perplexity Decider Flash v1.2 (4B) 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.

20 models · 21 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 → ~32GB 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 vision 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=32&max=any"to get this list in your terminal (add &format=json for agents).

Score vs release date

Release dates available for 7 of 20 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
1d1-ultra Liquid AI75±4api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)Etsabary · decision benchmarks
2d1.1 (hosted) Liquid AI74±4api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)Etsabary · decision benchmarks
3Jev 1.13 TypeSafe74±4api $0.042/M input tokensdocs.typesafe.ai ↗System One Bench (Ali Alhashimi)Etsabary · decision benchmarksBenchmark Heaven (own decision evaluations)
5Jev 1.14 TypeSafe71±4api $0.042/M input tokensdocs.typesafe.ai ↗System One Bench (Ali Alhashimi)Etsabary · decision benchmarksJevalKit (Valen Team)Benchmark Heaven (own decision evaluations)
6Mercury Decide (free endpoint) Inception70±4api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)
8Clef Cloudflare69±4api $0.24/M input tokens~68GBdevelopers.cloudflare.com ↗System One Bench (Ali Alhashimi)Etsabary · decision benchmarksBenchmark Heaven (own decision evaluations)
9Mercury Decide (paid endpoint) Inception68±7*api price unknownopenrouter.ai ↗Etsabary · decision benchmarks
13d1 (hosted) Liquid AI64±7*api price unknownopenrouter.ai ↗Etsabary · decision benchmarks
14Clef-flash · bf16 Cloudflare64±3local~25GBhow to run › System One Bench (Ali Alhashimi)Etsabary · decision benchmarksJevalKit (Valen Team)Benchmark Heaven (own decision evaluations)
17Span-02 (Gemma 4 26B-A4B) Respan63±3api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)Decision Index Vision 0.4
21Perplexity Decider Flash v1.2 (4B) · ollaya-f32 Perplexity62±3local~10GBhow to run › System One Bench (Ali Alhashimi)Decision Index Vision 0.4Etsabary · decision benchmarks
31GPT-6 Luna Decisions OpenAI58±3api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)Etsabary · decision benchmarks
35vLLM-SR d3-flash (9B) · ollaya-f32 vLLM Semantic Router55±5local~18GBhow to run › Decision Index Vision 0.4
40vLLM-SR d3-mini (4B) · ollaya-f32 vLLM Semantic Router51±5local~11GBhow to run › Decision Index Vision 0.4
48Clef-omni Cloudflare47±2api $0.15/M input tokens~87GBdevelopers.cloudflare.com ↗System One Bench (Ali Alhashimi)Decision Index Vision 0.4Etsabary · decision benchmarks
55baselabs joey-1 (2B) baselabs44±6*api price unknownopenrouter.ai ↗Etsabary · decision benchmarks
62vLLM-SR d3-nano (2B) · ollaya-f32 vLLM Semantic Router39±5local~8.2GBhow to run › Decision Index Vision 0.4
64LiquidAI d1-omni-600M · ollaya-f32 Liquid AI36±14*local~2.7GBhow to run › Benchmark Heaven (own decision evaluations)
75vLLM-SR d3-lite (0.8B) · ollaya-f32 vLLM Semantic Router24±5local~4.5GBhow to run › Decision Index Vision 0.4
79vLLM-SR d3-edge (0.6B) · ollaya-f32 vLLM Semantic Router10±5local~2.6GBhow to run › Decision Index Vision 0.4

+ 59 more don't fit your machine or budget · show them

79 of 79 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

20 models · 21 benchmarks · updated