whatmodel.sh / decision / Perplexity Decider Flash v1.2 (4B)

Perplexity Decider Flash v1.2 (4B)

+ contribute

Perplexity Decider Flash v1.2 (4B) is an open-weights decision model from Perplexity. Its published size is 4.54B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact Decision Index tested engine pplx-decider-flash-v1.2; reported model Perplexity Decider Flash v1.2 (4B). Type: full fine-tune; technique: autoregressive. No API price or speed inferred from the benchmark. Current repository metadata checked separately; an unpinned alias is not evidence that today’s weights reproduce this historical run. Reviewed Ollaya local route added 2026-10-31: pplx-decider:flash, manifest e403f9b1f0ac01129919433e754f66b8718ff76a61055eb841ec4f969173fd40, matching the independent accuracy-run manifest. Native family implementation; published runtime precision label F32. Memory is measured peak GPU increase (10622 MiB, rounded upward to 10.4 GiB) on RTX 5090 during the short five-question triage workload, not download size or a maximum-context guarantee. Runtime installation is required; no fresh execution performed here. Historical benchmark adapter and input protocol still vary by result.

Model information sources

· Weights licence ↗ · Parameter count ↗

Publisher parameter count; includes the served backbone, not just adapter size.

61.8±0.9#90 of 355 decision modelsfrom 92 benchmark results across 6 sources

$ how it compares

The decision models ranked either side of Perplexity Decider Flash v1.2 (4B), out of 357 with a score, then more from Perplexity.

#	model	score	where	RAM	released
88	Kev 27B Jared Palmer	62±1	model files		2026-09
89	Shisa DE-2.1 (repeat) Shisa.AI	62±1	model files
90	Perplexity Decider Flash v1.2 (4B) Perplexity	62±1	local	~10GB
91	Wald-26B-A4B ORG2 AI	62±1	model files
92	Blink v0.3 26B-A4B (NVFP4) Pixilab.ai	62±1	model files
	more from Perplexity
10	Perplexity Decider v1.1 (27B) Perplexity	72±1	api price unknown		2026-10-05
17	Perplexity Decider v1.2 (27B) Perplexity	70±1	model files

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

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #104 of 250 · 3.0% of its score · 6 tests

Beats Xor 26B-A4B v1.2, Senova-4B v2, SynACK Decide 26B-A4B v7 here, though they rank higher overall. Trails OneJev1.6-0.8B, TokenRhythm NeoHorse-Jev-4B, mmastrac diffgemma (decision readout) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: API-Bank (accuracy) submitted	86.4% ✎	▁▁▁▁▁▁▁▂▃▂█▂ beats 200/236	easy · avg 70%	▇ 0.9%	sharp · noisy
Decision Index 0.4.1: BFCL (case exact accuracy) submitted	95.8% ✎	▁▂▁▁▁▂▂▄▅▆█▁ beats 143/246	very easy · avg 92%	▇ 0.6%	sharp · maxed out · noisy
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	75.0% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 183/243	medium · avg 37%	▇ 0.5%	sharp · at its level · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	65.1% ✎	▁▁▁▁▁·▁▁▁▄█▁ beats 137/246	medium · avg 60%	▇ 0.5%	blunt · at its level
Decision Index 0.4.1: When2Call (accuracy) submitted	89.6% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 214/246	easy · avg 72%	▇ 0.5%	noisy
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	79.8% ✎	no other model yet		—	failed run: Publisher excludes this protocol because its prompt reveals GPT-4 as the best route. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json

retrieval & classification▁▁▁·▁▁▁▁▃▃█▅ #65 of 258 · 8.9% of its score · 13 tests

Beats AutoTrust-JEV-27B v1.1, oenogarum (Gemma 4 31B), Aleria Judge 1.1.1 (Gemma 4 31B) (decision readout) here, though they rank higher overall. Trails StartLux-Decision v1.1 0.8B, Standard One 8B SH (v4), StartLux-Decision v1.1 2B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
System One Bench: MIRACL pairwise (Arabic)	90.5% ✎	▂···▁··▃▄▂█▇ beats 22/29	easy · avg 84%	▇ 1.9%	typical
System One Bench: MASSIVE intent (Arabic)	86.0% ✎	▄·▁·▁▃▃▃█▄▇▇ beats 18/29	easy · avg 76%	▇ 1.7%	typical
Decision Index 0.4.1: BRIGHT (nDCG@10) submitted	45.6% ✎	▁▁▁▁▁▁▁▂▂▄▇█ beats 164/246	medium · avg 38%	▇ 1.3%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	57.8% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 188/246	medium · avg 47%	▇ 1.2%	at its level
System One Bench: MIRACL rerank (Arabic)	88.5% ✎	▁▂·▁▁·▂·▄▄█▇ beats 22/29	easy · avg 79%	▇ 1.1%	typical
System One Bench: MASSIVE scenario (Arabic)	70.5% ✎	▃▁·▄▃▂█▂▁▁·▃ beats 15/29	medium · avg 60%	▇ 0.7%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	86.2% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 196/246	easy · avg 74%	▇ 0.3%	blunt
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	72.5% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 141/245	easy · avg 70%	▇ 0.3%	noisy

5 more tests

test	result	vs the others in its run	difficulty	counts	why
System One Bench: AG News (English)	89.0% ✎	▁···▅▁▅▂▆█▂▅ beats 14/29	easy · avg 88%	▇ 0.2%	blunt
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	93.3% ✎	▁▁▁▁▁▁▁▂▃▆█▄ beats 181/238	easy · avg 83%	▇ 0.2%	sharp · noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	90.1% ✎	▁·▁▁▁▁▁▁▂█▇█ beats 181/238	easy · avg 79%	▇ 0.1%	noisy
System One Bench: Banking77 (English)	90.5% ✎	█▅▆█▅▃▃▃···▂ beats 17/26	very easy · avg 87%	▇ 0.0%	blunt · noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	35.7% ✎	no other model yet		—	failed run: Publisher reports a builder bug exposing dialogue turns after the classified turn; awaiting corrected rerun. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json

language understanding▂▂▂▃▂▅▆▃█▆▄▃ #78 of 256 · 15% of its score · 18 tests

Beats Turion Jev 31B, AJev lm6q (Gemma 4 31B), oenogarum (Gemma 4 31B) here, though they rank higher overall. Trails Sieve 4B, Amber 4B, Decider 35B-A3B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: ANLI (macro-F1) submitted	69.3% ✎	▁▂▃▅▅█▂····▁ beats 151/246	medium · avg 60%	▇ 2.5%	at its level
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	82.3% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 178/246	easy · avg 74%	▇ 2.4%	typical
Decision Index 0.4.1: HellaSwag (accuracy) submitted	93.8% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 169/246	very easy · avg 85%	▇ 1.7%	sharp · maxed out
System One Bench: BoolQ (English)	89.0% ✎	▁▁▁▃·▁▃▃▇▇█▅ beats 12/29	easy · avg 86%	▇ 1.3%	typical
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	59.4% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 180/246	medium · avg 44%	▇ 1.2%	at its level
System One Bench: OSACT4 offensive (Arabic)	90.5% ✎	▁······▁▃█▇▄ beats 16/29	easy · avg 84%	▇ 1.2%	sharp
System One Bench: SST-2 (English)	95.0% ✎	▁·▃▁▂▂▃▅█·▂▁ beats 15/29	easy · avg 92%	▇ 0.8%	typical
Decision Index 0.4.1: ACOS (per-review F1) submitted	35.5% ✎	▁··▂▃▃▅▇█▇▃▃ beats 184/246	hard · avg 22%	▇ 0.7%	at its level · noisy

10 more tests

test	result	vs the others in its run	difficulty	counts	why
System One Bench: SST-5 (English)	54.0% ✎	▃·▁▁▃·▇▄▄▄█▅ beats 13/29	medium · avg 51%	▇ 0.7%	blunt · at its level
Decision Index 0.4.1: VAST (macro-F1) submitted	76.9% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 180/246	medium · avg 62%	▇ 0.5%	at its level
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	82.9% ✎	▁▁▁▁▁▁▁▂▂▇▇█ beats 194/246	easy · avg 71%	▇ 0.5%	noisy
System One Bench: XNLI Arabic (Arabic)	85.3% ✎	▄▁▃▁▄▇█▅▃▃·▃ beats 27/29	medium · avg 68%	▇ 0.4%	at its level
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	82.5% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 205/246	easy · avg 66%	▇ 0.4%	noisy
System One Bench: XNLI French (French)	88.0% ✎	▂▅·▁▁▅▅█▃▃·▃ beats 26/29	medium · avg 76%	▇ 0.3%	blunt · at its level
System One Bench: Tweet sentiment (Arabic)	71.5% ✎	▁······▁▁▄█▄ beats 22/29	medium · avg 62%	▇ 0.3%	blunt · at its level
Decision Index 0.4.1: FinEntity (macro-F1) submitted	90.5% ✎	▁···▁▁▁▁▃▆█▃ beats 152/246	easy · avg 84%	▇ 0.3%	noisy
System One Bench: XNLI English (English)	92.7% ✎	▃·▁·▃▇█▄█▃▃▁ beats 26/29	easy · avg 85%	▇ 0.1%	blunt · noisy
System One Bench: DAIR Emotion (English)	57.0% ✎	██▃·····▁··▁ beats 15/29	medium · avg 59%	▇ 0.0%	blunt · at its level

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #87 of 247 · 4.6% of its score · 6 tests

Beats Jev 1.13, jev-brain · Gemma-4-31B (decision readout), Jeffan-4B-2010 here, though they rank higher overall. Trails Amber 0.4B, Crivo-1B, GLiClass v4 edge (0.4B) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: New Yorker (accuracy) submitted	74.1% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 180/246	medium · avg 61%	▇ 1.9%	at its level
Decision Index 0.4.1: Humicroedit (accuracy) submitted	61.6% ✎	▁·▁▃▂▂▅█▄▂▁▁ beats 135/246	medium · avg 60%	▇ 1.0%	blunt · at its level
Decision Index 0.4.1: cfcolor (accuracy) submitted	66.2% ✎	▁········▁▇█ beats 212/246	medium · avg 60%	▇ 0.6%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	27.2% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 145/237	hard · avg 21%	▇ 0.6%	sharp · at its level · noisy
Decision Index 0.4.1: BPoMP (accuracy) submitted	91.3% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 158/246	easy · avg 83%	▇ 0.5%	sharp · noisy
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	61.6% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 217/246	medium · avg 47%	▇ 0.2%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▂ #99 of 263 · 34% of its score · 21 tests

Beats KnowLine 4B Gen5.5, Jeffan-4B-2010, Senova-4B v2 here, though they rank higher overall. Trails Decider 35B-A3B, decisio 0.4.0 (Qwen3.6-35B-A3B) (decision readout), Solar Decide here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Etsabary · 1342 text decisions	68.8% ✎	▁··▂·▃▅▁▃▂▂█ beats 8/24	medium · avg 47%	▇▇▇ 8.7%	sharp · at its level
Decision Index 0.4.1: MMLU-Pro (accuracy) submitted	59.1% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 150/246	medium · avg 46%	▇▇ 5.2%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	80.1% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 158/246	easy · avg 70%	▇ 2.9%	sharp
Decision Index 0.4.1: BBH (accuracy) submitted	73.4% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 152/246	easy · avg 66%	▇ 2.8%	typical
CONFLICTS · source conflict types	61.8% ✎	▃▃··▅▃···▃▃█ beats 2/9	medium · avg 45%	▇ 2.2%	at its level
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	67.6% ✎	▁··▄█▆▆▆▅▅▃▁ beats 180/246	medium · avg 45%	▇ 1.7%	sharp · at its level
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	44.4% ✎	▂▆▅▅▆█▆▃▁··▁ beats 148/246	medium · avg 39%	▇ 1.7%	blunt · at its level
Decision Index 0.4.1: WinoGrande (accuracy) submitted	88.6% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 191/246	easy · avg 78%	▇ 1.3%	typical

13 more tests

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	95.7% ✎	▂▂▂▂▂▂▁▃▄▆█▆ beats 154/246	very easy · avg 90%	▇ 1.2%	sharp · maxed out
Ollaya · PubMedQA	77.6% ✎	▃▂▃▃·▂▃▃▆█▆▂ beats 23/25	easy · avg 69%	▇ 1.0%	typical
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	11.7% ✎	▁▁▁▁▁▁▃▂▅█▂▁ beats 50/246	hard · avg 22%	▇ 0.8%	at its level
Ollaya · BoolQ	88.0% ✎	▁··▃▃▁▄▁▃█▇▄ beats 22/25	easy · avg 84%	▇ 0.7%	typical
Decision Index 0.4.1: MuSR (accuracy) submitted	60.2% ✎	▁▁▂█▅▄▁▁▁▁▁▁ beats 135/246	medium · avg 60%	▇ 0.6%	blunt · at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	62.6% ✎	██▄▃▃▄▁▁···▁ beats 149/246	medium · avg 58%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	99.1% ✎	▁▁▂▁▂▂▂▂▃▆█▅ beats 206/246	very easy · avg 96%	▇ 0.5%	sharp · maxed out · noisy
Decision Index 0.4.1: CLadder (accuracy) submitted	85.4% ✎	▁███▆▅▃▃▁▁▁▁ beats 205/246	easy · avg 71%	▇ 0.5%	typical
Ollaya · SQuAD 2 answerability	92.6% ✎	▁▁▃▁▇█▃▃▄▃·▃ beats 25/25	easy · avg 80%	▇ 0.4%	blunt
Ollaya · VitaminC verification	90.8% ✎	▁···▃·▄█▃··▃ beats 25/25	easy · avg 78%	▇ 0.4%	blunt
Decision Index 0.4.1: ChessBench (accuracy) submitted	21.2% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 173/246	hard · avg 16%	▇ 0.4%	blunt
Ollaya · MultiNLI	92.0% ✎	▂··▄██▂▆▆█▆▂ beats 24/25	very easy · avg 87%	▇ 0.2%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	12.6% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 131/246	very hard · avg 13%	▇ 0.0%	blunt

vision▁·▃▂▃▃▇▄▄█▄▃ #21 of 79 · 22% of its score · 18 tests

Beats StartLux-Decision v1.1 27B, Nace.AI Drex Max v1 (27B), vLLM-SR d3 (27B) here, though they rank higher overall. Trails Clef-flash here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Etsabary · 200 image decisions	64.5% ✎	▂▂··▂▃▂·█▃▂▂ beats 4/14	medium · avg 45%	▇▇ 5.9%	sharp · at its level
System One Bench: Image object count	87.5% ✎	▃····▃█▃▅·█▃ beats 2/11	easy · avg 71%	▇ 2.7%	sharp
System One Bench: SST-2 screenshot	93.5% ✎	█···███·█·▄▄ beats 4/11	easy · avg 87%	▇ 2.1%	sharp
System One Bench: Visual yes/no questions	87.5% ✎	▂······▃█▃·▃ beats 6/11	medium · avg 80%	▇ 1.8%	at its level
System One Bench: Oxford pets classification	84.0% ✎	███·█▄▄·▄··▄ beats 2/11	easy · avg 77%	▇ 1.6%	typical
System One Bench: Arabic sentiment screenshot	35.0% ✎	▄··▄··▂···▆█ beats 2/11	very hard · avg 37%	▇ 1.3%	sharp · noisy
Decision Index Vision 0.4 · Mind2Web (chance-corrected points) submitted	92.8700 points ✎	▁·▁▁▃▁▄▄▃▅█▃ beats 59/61	avg 63.423 points	▇ 0.8%	blunt
Decision Index Vision 0.4 · MMMU-Pro vision (chance-corrected points) submitted	31.5100 points ✎	▁▁▅▁▄▁▄█▄▃▇▇ beats 32/61	avg 22.797 points	▇ 0.8%	blunt

10 more tests

test	result	vs the others in its run	difficulty	counts	why
Decision Index Vision 0.4 · Winoground (chance-corrected points) submitted	83.5000 points ✎	▁·▁▁▁▁·▂▁▃▅█ beats 42/61	avg 66.987 points	▇ 0.7%	blunt
Decision Index Vision 0.4: private same skills (points)	59.4800 points ✎	▁▁▃▃▁▂▄▆▃▃▅█ beats 39/61	avg 48.067 points	▇ 0.7%	blunt
Decision Index Vision 0.4 · Moderation (Hateful Memes) (chance-corrected points) submitted	54.4900 points ✎	▆▃▃▆▄▄▇██▁▆▆ beats 57/61	avg 28.443 points	▇ 0.6%	blunt
Decision Index Vision 0.4 · RealWorldQA (chance-corrected points) submitted	63.1000 points ✎	▁·▂▂▁▃▃▅▅▃▂█ beats 37/61	avg 53.651 points	▇ 0.6%	blunt
Decision Index Vision 0.4 · CharXiv (chance-corrected points) submitted	76.9400 points ✎	▂▁▁▁▁▃▁▅▅▆██ beats 41/61	avg 62.137 points	▇ 0.5%	blunt
Decision Index Vision 0.4 · InfographicVQA (chance-corrected points) submitted	93.9900 points ✎	▁·▁▂▁▁▃▁▂▂▅█ beats 35/61	avg 83.859 points	▇ 0.4%	blunt
Decision Index Vision 0.4 · R-Bench-M (chance-corrected points) submitted	23.4800 points ✎	▂▄▇▆█▄▄▇▂··▁ beats 30/61	avg 19.676 points	▇ 0.4%	blunt
Decision Index Vision 0.4 · BLINK (chance-corrected points) submitted	55.5400 points ✎	▁···▁▂▂▂█▄▂▃ beats 37/61	avg 44.994 points	▇ 0.4%	blunt
Decision Index Vision 0.4 · CV-Bench (chance-corrected points) submitted	75.6400 points ✎	▁▁▁▁▂▁▁▁▄█▃▃ beats 34/61	avg 67.662 points	▇ 0.3%	blunt
Decision Index Vision 0.4 · KIE (CORD+FUNSD) (chance-corrected points) submitted	91.7200 points ✎	▁··▁▁▁▁▁▂▄█▁ beats 33/61	avg 85.184 points	▇ 0.2%	blunt

workflows▁▁▁▂▁▂▂▆█▄▁▁ #26 of 114 · 7.9% of its score · 10 tests

Beats d1.1 (hosted), Span-02 (Gemma 4 26B-A4B), Mercury Decide (free endpoint) here, though they rank higher overall. Trails imajev-4b (DecisionBench c9e5f132), reflex-qwen3.5-4b-lora (DecisionBench 9df1cbbb), litjev (DecisionBench 1d4bf0f2) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Ollaya · Typed decisions test	72.6% ✎	▄·▂▆▄▄▂·▄█▆▆ beats 21/22	medium · avg 64%	▇ 1.7%	at its level
System One Bench: Typed decisions (English)	73.0% ✎	▁▁·▁▁▁▅▄▇██▁ beats 21/29	medium · avg 63%	▇ 1.7%	blunt · at its level
Ollaya · SummEval relevance	54.2% ✎	▁··▁▂▁▅··▃██ beats 25/25	medium · avg 41%	▇ 1.1%	at its level
Ollaya · PAWS paraphrase	93.6% ✎	▂▂▂▂·▃█▆·▂·▃ beats 25/25	easy · avg 81%	▇ 0.9%	typical
Ollaya · Aegis 2 safety	84.4% ✎	▂·▁▁▂·▃▃▂█▃▁ beats 24/25	easy · avg 77%	▇ 0.8%	typical
Ollaya · MASSIVE German intent	91.7% ✎	▁····▁·▂▃▆█▃ beats 23/23	easy · avg 84%	▇ 0.6%	sharp
Ollaya · HelpSteer 2 scoring	38.6% ✎	▁▁··▁·▁▃▇█▅▄ beats 14/23	medium · avg 38%	▇ 0.4%	blunt · at its level
Ollaya · SummEval consistency	84.0% ✎	▁·▁▁··▁▁▁▁▄█ beats 20/25	easy · avg 75%	▇ 0.4%	sharp · noisy

2 more tests

test	result	vs the others in its run	difficulty	counts	why
Ollaya · MASSIVE English intent	91.4% ✎	▁·······▂▃█▃ beats 23/23	very easy · avg 86%	▇ 0.3%	maxed out · noisy
Ollaya · Civil Comments	92.7% ✎	▂▂▂·▂▅▅▆▃▅·█ beats 20/23	very easy · avg 86%	▇ 0.1%	blunt · noisy

other tests2 not in a task · 4.4% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	55.7600 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 175/247	avg 41.206 points	▇ 2.8%	blunt
Decision Index 0.4.1: private new domains (equated points)	52.0700 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 164/247	avg 38.861 points	▇ 1.6%	blunt

$ where to run it

Other ways to run it

Run it as published · needs 10.4 GB, NVIDIA only · from github.com/ollaya-dev
ollaya run pplx-decider:flash --preset triage "The customer requests a refund."

Download model files ↗
Published model files verified. This is a verified repository snapshot; the benchmark does not establish an identical weight revision. A matching Ollaya runtime manifest and published short-workload GPU memory measurement were reviewed separately; see local variants and identity_note. The command was not executed in this review. Source · checked 2026-10-31.

Model repository and documentation: huggingface.co/perplexity-ai ↗

$ how the score was worked out · Decision Index 0.4.1 46%, System One Bench (Ali Alhashi… 22%, Etsabary · decision benchmarks 15% and 3 more

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

source	results	share of the score	licence
Decision Index 0.4.1 · submitted; independent execution unverified	42	▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 46%	CC BY 4.0
System One Bench (Ali Alhashimi)	21	▇▇▇▇▇▇ 22%	CC-BY-4.0
Etsabary · decision benchmarks	2	▇▇▇▇ 15%	MIT
Ollaya · original runs	14	▇▇▇ 9%	Apache-2.0
Decision Index Vision 0.4 · submitted; independent execution unverified	12	▇▇ 7%	CC BY 4.0
Do We Need an LLM to Know Why Sources Conflict? Decision Models for Knowledge-Conflict Typing in RAG	1	▇ 2%	CC-BY-4.0
prior · pull toward the middle		▇ 0%
weighted average		= 61.8

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	79.8%	failed run: Publisher excludes this protocol because its prompt reveals GPT-4 as the best route. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json
Decision Index 0.4.1	Decision Index 0.4.1: SGD/SGD-X (macro-F1) [correct]	35.7%	failed run: Publisher reports a builder bug exposing dialogue turns after the classified turn; awaiting corrected rerun. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit 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 2026-11-03 02:37 UTC · sources and licences · json