{"id":"pplx-decider-v1-1-27b","slug":"pplx-decider-v1-1-27b","name":"Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)","orgId":"perplexity","family":"pplx-decider-v1-1-27b","releaseDate":null,"availability":"ga","openWeights":true,"license":"apache-2.0","paramsTotal":27800000000,"contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://huggingface.co/perplexity-ai/pplx-decider-v1.1-27b","hfRepo":"perplexity-ai/pplx-decider-v1.1-27b","modalities":["text"],"aliases":["Perplexity Decider v1.1 (27B)","Perplexity Decider V1.1","Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)","Perplexity Decider V1.1 27B","pplx-decider-v1.1-27b","pplx-decider-v1-1-27b"],"aliasNote":null,"decision":{"kind":"decision","outputTypes":["noul","choice","score","multi"],"vision":false,"backbone":"Qwen3.8-27B","technique":"autoregressive","weightsRepo":null,"servedParams":27800000000,"hosted":null,"coverage":0.9993,"partialRun":true,"latencyMedianMs":104.1,"latencyHosted":false,"firstSeen":"2026-10-08","reliability":[{"w":0.002,"acc":0.0484,"conf":0.0824},{"w":0.0176,"acc":0.2082,"conf":0.1542},{"w":0.0218,"acc":0.2419,"conf":0.2557},{"w":0.0257,"acc":0.376,"conf":0.3553},{"w":0.0406,"acc":0.4457,"conf":0.4517},{"w":0.0702,"acc":0.4576,"conf":0.5465},{"w":0.0619,"acc":0.5174,"conf":0.6473},{"w":0.0805,"acc":0.6059,"conf":0.7517},{"w":0.1027,"acc":0.6701,"conf":0.8547},{"w":0.5771,"acc":0.9392,"conf":0.9811}],"priceUsdPer1k":0.21732654028436021,"api":false,"measuredOn":"2026-10-06"},"paramsSource":"huggingface","organization":{"id":"perplexity","name":"Perplexity","slug":"perplexity","country":null,"url":null,"color":"#A39E8F","mark":"PE"},"price":null,"answer":{"word":"Yes.","scope":"SOTA in Decision models (tied).","lead":"As of Oct 15, 2026, Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) is tied for state of the art with low confidence in Decision models (since Oct 14, 2026; sources Decision Index Full and JevBench; methodology v1.0). Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) is co-SOTA in Decision models with low confidence because primary sources disagree beyond the margin of error (with vLLM-SR d3 (27B), linne-oak 1.0 (31B)).","confidence":"low"},"verdicts":[{"id":"v_overall_decision_pplx-decider-v1-1-27b_2026-10-14_open_CO_SOTA","model":"pplx-decider-v1-1-27b","model_name":"Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)","category":"decision","category_name":"Decision models","slice":"overall","state":"CO_SOTA","rank":1,"present_evidence":[],"missing_evidence":[],"coverage":null,"segments":[],"confirm":[],"margin":0,"confidence":"low","valid_from":"2026-10-14","valid_to":null,"configuration":null,"rationale":"Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) is co-SOTA in Decision models with low confidence because primary sources disagree beyond the margin of error (with vLLM-SR d3 (27B), linne-oak 1.0 (31B)).","safeguard_flags":["source_conflict"],"co_sota_with":["vllm-sr-d3-v310","linne-oak-1-0"],"challengers":[],"methodology":"1.0","label":"Co-SOTA · Decision models"},{"id":"v_open-weights_decision_pplx-decider-v1-1-27b_2026-10-14_open_CO_SOTA","model":"pplx-decider-v1-1-27b","model_name":"Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)","category":"decision","category_name":"Decision models","slice":"open-weights","state":"CO_SOTA","rank":1,"present_evidence":[],"missing_evidence":[],"coverage":null,"segments":[],"confirm":[],"margin":0,"confidence":"low","valid_from":"2026-10-14","valid_to":null,"configuration":null,"rationale":"Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) is co-SOTA in Decision models with low confidence because primary sources disagree beyond the margin of error (with vLLM-SR d3 (27B), linne-oak 1.0 (31B)).","safeguard_flags":["source_conflict"],"co_sota_with":["vllm-sr-d3-v310","linne-oak-1-0"],"challengers":[],"methodology":"1.0","label":"Co-SOTA · Decision models (open weights)"},{"id":"v_le-32b_decision_pplx-decider-v1-1-27b_2026-10-14_open_CO_SOTA","model":"pplx-decider-v1-1-27b","model_name":"Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)","category":"decision","category_name":"Decision models","slice":"le-32b","state":"CO_SOTA","rank":1,"present_evidence":[],"missing_evidence":[],"coverage":null,"segments":[],"confirm":[],"margin":0,"confidence":"low","valid_from":"2026-10-14","valid_to":null,"configuration":null,"rationale":"Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) is co-SOTA in Decision models with low confidence because primary sources disagree beyond the margin of error (with vLLM-SR d3 (27B), linne-oak 1.0 (31B)).","safeguard_flags":["source_conflict"],"co_sota_with":["vllm-sr-d3-v310","linne-oak-1-0"],"challengers":[],"methodology":"1.0","label":"Co-SOTA · Decision models (≤32B)"},{"id":"v_latency_decision_pplx-decider-v1-1-27b_2026-10-15_open_RANKED","model":"pplx-decider-v1-1-27b","model_name":"Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)","category":"decision","category_name":"Decision models","slice":"latency","state":"RANKED","rank":5,"present_evidence":[],"missing_evidence":[],"coverage":null,"segments":[],"confirm":[],"margin":40.3,"confidence":"medium","valid_from":"2026-10-15","valid_to":null,"configuration":null,"rationale":"Ranked #5 in Decision models on same-GPU latency. Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) is inside the field.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#5 in Decision models (same-GPU latency)"},{"id":"v_price_decision_pplx-decider-v1-1-27b_2026-10-15_open_RANKED","model":"pplx-decider-v1-1-27b","model_name":"Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout)","category":"decision","category_name":"Decision models","slice":"price","state":"RANKED","rank":4,"present_evidence":[],"missing_evidence":[],"coverage":null,"segments":[],"confirm":[],"margin":0.17647989844278947,"confidence":"medium","valid_from":"2026-10-15","valid_to":null,"configuration":null,"rationale":"Ranked #4 in Decision models on price per 1,000 decisions. Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) is inside the field.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#4 in Decision models (price per 1,000 decisions)"}]}