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

Best Decision models for tools & automation to run locally

A leaderboard of the best Decision models for tools & automation you can run on your own computer. Clef-flash is #1 right now, followed by Clef-omni and Clef. Published typed-decision evaluations, with evaluator-specific protocols and tested settings retained. Each task score uses only its reviewed benchmarks. Each model shows the download that fits and the command to run it.

25 models · 9 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 · local only · reset
ram[2][4][8][16][24][32][48][64][96][128][192][256]
[]
GB
gpu[apple silicon][graphics card][unified memory][none]
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 tools & automation evaluations only. Compare scores within this task. Models without task measurements are excluded.

New #1s:
orRSS

detecting your OS…

$ terminal

detecting your OS…

Run curl "whatmodel.sh/decision?max=0"to get this list in your terminal (add &format=json for agents).

Score vs release date

Release dates available for 9 of 25 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
1Clef-flash Cloudflare70±2local~25GBhow to run › Decision Index 0.4.1AgentDecisionBenchDecideBench (Cho Yin Yong)
2Clef-omni Cloudflare70±2local~87GBhow to run › Decision Index 0.4.1
3Clef Cloudflare69±1local~68GBhow to run › Decision Index 0.4.1DecideBench (Cho Yin Yong)
45Bespoke Nimble 9B v3 Bespoke Labs63±1†local~21GBhow to run › Decision Index 0.4.1
62Jebadiah 27B v2.1 (T1e) frontier-infra62±1†local~69GBhow to run › Decision Index 0.4.1
91vLLM-SR d3-flash (9B) vLLM Semantic Router59±1†local~18GBhow to run › Decision Index 0.4.1
104Perplexity Decider Flash v1.2 (4B) Perplexity57±1†local~10GBhow to run › Decision Index 0.4.1
110vLLM-SR d3-mini (4B) vLLM Semantic Router57±1†local~11GBhow to run › Decision Index 0.4.1
128Jebadiah 9B v2.1 (T1e) frontier-infra54±1†local~25GBhow to run › Decision Index 0.4.1
133vLLM-SR Decision 2.0 Lux 9B vLLM Semantic Router53±1†local~20GBhow to run › Decision Index 0.4.1
141Kev 4B r10 Jared Palmer51±1†local~11GBhow to run › Decision Index 0.4.1
156vLLM-SR d3-nano (2B) vLLM Semantic Router50±1†local~8.2GBhow to run › Decision Index 0.4.1
164Bespoke Nimble 9B v2 Bespoke Labs49±1†local~20GBhow to run › Decision Index 0.4.1
170vLLM-SR Decision 2.0 Nox 4B vLLM Semantic Router48±1†local~12GBhow to run › Decision Index 0.4.1
173GLiClass v4 edge (0.4B) Knowledgator Engineering47±1†local~2.7GBhow to run › Decision Index 0.4.1
183Decider 4B Mapika44±1†local~11GBhow to run › Decision Index 0.4.1
188vLLM-SR d3-lite (0.8B) vLLM Semantic Router42±1†local~4.5GBhow to run › Decision Index 0.4.1
200vLLM-SR Decision 2.0 Sol 2B vLLM Semantic Router39±1†local~11GBhow to run › Decision Index 0.4.1
202Strands Decider 2B (Hobson v21) Amazon Web Services38±1local~6.6GBhow to run › Decision Index 0.4.1
214vLLM-SR d3-edge (0.6B) vLLM Semantic Router33±0†local~2.6GBhow to run › Decision Index 0.4.1
216Kev 0.8B r15 Jared Palmer31±0†local~4.6GBhow to run › Decision Index 0.4.1
225vLLM-SR Decision 2.0 Eos 0.8B vLLM Semantic Router25±0†local~5.8GBhow to run › Decision Index 0.4.1
228LiquidAI d1-omni-600M Liquid AI20±0†local~2.7GBhow to run › Decision Index 0.4.1
233vLLM-SR Decision 2.0 Kai 0.6B vLLM Semantic Router18±0†local~3.7GBhow to run › Decision Index 0.4.1
237CLM-v0.1-8B Qwen3-8B heads Contrastive-LM16±0local~15GBhow to run › Decision Index 0.4.1DecideBench (Cho Yin Yong)AgentDecisionBench

+ 225 more don't fit your machine or budget · show them

250 of 250 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

25 models · 9 benchmarks · updated