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

Best Decision models for retrieval & classification for a 16GB graphics card

A leaderboard of the best Decision models for retrieval & classification for a 16GB graphics card (with 32GB of system RAM). The best one that runs on it right now is Bespoke Nimble 9B v3, 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.

38 models · 22 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 → ~16GB 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 retrieval & classification 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=16&max=any"to get this list in your terminal (add &format=json for agents).

Score vs release date

Release dates available for 15 of 38 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
35Microsoft-Decision-1 Microsoft64±3api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)LeanMCP · JevBench
38d1 (hosted) Liquid AI63±3api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)LeanMCP · JevBench
41Perplexity Decider v1.1 (27B) Perplexity62±2api price unknownopenrouter.ai ↗DecisionPivot · StartLux LabsLeanMCP · JevBenchCitead GEO Citation Benchmark (October 2026)
46d1-ultra Liquid AI62±3api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)
58Span-02 (Gemma 4 26B-A4B) Respan61±1api price unknownopenrouter.ai ↗Decision Index 0.4.1System One Bench (Ali Alhashimi)
60d1.1 (hosted) Liquid AI61±3api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)
61Clef Cloudflare61±1api $0.24/M input tokens~68GBdevelopers.cloudflare.com ↗Decision Index 0.4.1DecisionPivot · StartLux LabsSystem One Bench (Ali Alhashimi)LeanMCP · JevBenchDecideBench (Cho Yin Yong)
62Clef-omni Cloudflare61±1api $0.15/M input tokens~87GBdevelopers.cloudflare.com ↗Decision Index 0.4.1System One Bench (Ali Alhashimi)LeanMCP · JevBench
63Bespoke Nimble 9B v3 · ollaya-f32 Bespoke Labs61±1†local (slow)~21GBhow to run › Decision Index 0.4.1
65Perplexity Decider Flash v1.2 (4B) · ollaya-f32 Perplexity61±1local~10GBhow to run › Decision Index 0.4.1System One Bench (Ali Alhashimi)
68Clef-flash Cloudflare60±1api $0.038/M input tokens +local~25GBdevelopers.cloudflare.com ↗Decision Index 0.4.1DecisionPivot · StartLux LabsSystem One Bench (Ali Alhashimi)LeanMCP · JevBenchDecideBench (Cho Yin Yong)
77Mercury Decide (free endpoint) Inception59±3api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)
83Mercury Decide (paid endpoint) Inception59±6*api price unknownopenrouter.ai ↗LeanMCP · JevBench
93Jev 1.13 TypeSafe58±1api $0.042/M input tokensdocs.typesafe.ai ↗Decision Index 0.4.1System One Bench (Ali Alhashimi)DecisionPivot · StartLux LabsLeanMCP · JevBenchCitead GEO Citation Benchmark (October 2026)DecideBench (Cho Yin Yong)
106Jev 1.14 TypeSafe57±3api $0.042/M input tokensdocs.typesafe.ai ↗System One Bench (Ali Alhashimi)
111baselabs joey-1 (2B) baselabs57±1api price unknownopenrouter.ai ↗Decision Index 0.4.1System One Bench (Ali Alhashimi)
115vLLM-SR d3-flash (9B) · ollaya-f32 vLLM Semantic Router57±1†local (slow)~18GBhow to run › Decision Index 0.4.1
118GPT-6 Luna Decisions OpenAI56±3api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)LeanMCP · JevBenchDecideBench (Cho Yin Yong)
129Jebadiah 9B v2.1 (T1e) · bf16-t1e frontier-infra55±1†local (slow)~25GBhow to run › Decision Index 0.4.1
135vLLM-SR d3-mini (4B) · ollaya-f32 vLLM Semantic Router55±1†local~11GBhow to run › Decision Index 0.4.1
141GLiClass v4 edge (0.4B) · ollaya-f32 Knowledgator Engineering55±1†local~2.7GBhow to run › Decision Index 0.4.1
161vLLM-SR d3-nano (2B) · ollaya-f32 vLLM Semantic Router50±1local~8.2GBhow to run › Decision Index 0.4.1System One Bench (Ali Alhashimi)
168vLLM-SR Decision 2.0 Nox 4B · ollaya-f32 vLLM Semantic Router49±1†local~12GBhow to run › Decision Index 0.4.1
173vLLM-SR Decision 2.0 Lux 9B · ollaya-f32 vLLM Semantic Router48±1†local (slow)~20GBhow to run › Decision Index 0.4.1
174Bespoke Nimble 9B v2 · ollaya-f32 Bespoke Labs48±1†local (slow)~20GBhow to run › Decision Index 0.4.1
179Solar Decide Upstage47±2api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)LeanMCP · JevBench
183Decider 4B · ollaya-f32 Mapika45±1local~11GBhow to run › Decision Index 0.4.1Citead GEO Citation Benchmark (October 2026)
185vLLM-SR d3-lite (0.8B) · ollaya-f32 vLLM Semantic Router45±1local~4.5GBhow to run › Decision Index 0.4.1System One Bench (Ali Alhashimi)
192Kev 4B r10 · ollaya-f32 Jared Palmer44±1local~11GBhow to run › Decision Index 0.4.1System One Bench (Ali Alhashimi)
203Solar Decide Flash Upstage41±2api price unknownopenrouter.ai ↗System One Bench (Ali Alhashimi)LeanMCP · JevBench
212vLLM-SR Decision 2.0 Eos 0.8B · ollaya-f32 vLLM Semantic Router37±2†local~5.8GBhow to run › Decision Index 0.4.1
214vLLM-SR Decision 2.0 Sol 2B · ollaya-f32 vLLM Semantic Router37±2†local~11GBhow to run › Decision Index 0.4.1
223vLLM-SR d3-edge (0.6B) · ollaya-f32 vLLM Semantic Router32±1local~2.6GBhow to run › Decision Index 0.4.1System One Bench (Ali Alhashimi)
228vLLM-SR Decision 2.0 Kai 0.6B · ollaya-f32 vLLM Semantic Router30±2†local~3.7GBhow to run › Decision Index 0.4.1
235Strands Decider 2B (Hobson v21) · bf16 Amazon Web Services27±2local~6.6GBhow to run › Decision Index 0.4.1
236LiquidAI d1-omni-600M · ollaya-f32 Liquid AI27±1local~2.7GBhow to run › Decision Index 0.4.1System One Bench (Ali Alhashimi)
241Kev 0.8B r15 · ollaya-f32 Jared Palmer24±1local~4.6GBhow to run › Decision Index 0.4.1System One Bench (Ali Alhashimi)
258CLM-v0.1-8B Qwen3-8B heads · ollaya-f32 Contrastive-LM-14±3local~15GBhow to run › Decision Index 0.4.1DecideBench (Cho Yin Yong)

+ 220 more don't fit your machine or budget · show them

258 of 258 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

38 models · 22 benchmarks · updated