AgentJev-0.6B: a small trained decision model that takes unstructured state (diffs, traces, logs) and returns calibrated distributions in one ~50 ms forward pass. Apache 2.0.

github.comPythonApache-2.0pushed oct 1334 forksadded sep 23

Qwen3.5 fine-tunes (0.8B up to a 35B MoE, plus a vision build), Apache 2.0 weights on Hugging Face. The vision build ranks #2 on Image JevBench. Serves TypeSafe /v1/systemone, so the official SDK works by repointing TYPESAFE_BASE_URL.

github.comPythonApache-2.0pushed oct 1249 forksadded sep 21

Nace.AI's open decision models: v1.5 on a distilled Qwen3.5-9B and DLM on an Efficient-DLM-8B diffusion backbone, each serving /v1/systemone three ways: Python/vLLM, a llama.cpp GGUF fork, or Ollama.

github.comPythonApache-2.0pushed oct 133 forksadded oct 12

SberAI's Russian-first encoder decision model: choice / score / noul plus a ranking question type, all questions scored in one packed encoder pass with an exact state cache. Torch, int8 ONNX, MLX, or vLLM.

github.comPythonpushed oct 102 forksadded oct 12

HIT-TMG's recipe for turning an embedding model into a decision model: choice / score / noul heads over your pick of embedder, with a Hugging Face model collection and dataset. Apache 2.0.

github.comPythonApache-2.0pushed oct 125 forksadded oct 13

Open-weight Jev alternative (4B and 9B, Apache 2.0) serving /v1/systemone at about 13 ms on an H100. JevBench v1.4 ranked v0.2 second of 76 systems and first open entrant, though the project itself reports Jev edging it on 308 fresh sealed items. Two-teacher distillation, pre-registered epoch sweeps.

github.comPythonApache-2.0pushed sep 2812 forksadded oct 13

Train a small one-pass scorer over a changing list of text options. Each option queries the context; a shared head returns one probability per option. Byte encoder from scratch, or a frozen Hugging Face encoder.

github.comPythonMITpushed sep 16120 forksadded sep 20

MiniCPM5-1B cut to 17 layers with a one-logit head, quantized to GGUF and run on CPU via llama.cpp. Serves TypeSafe /v1/systemone for noul (yes/no) questions only, so the official SDK works by repointing TYPESAFE_BASE_URL. MIT.

github.comC++MITpushed oct 12154 forksadded sep 27

Small trained decision models on Qwen3.5 (0.8B and 4B) serving /v1/systemone at ~50 ms on one H100. jiwo-0.8b ranks #1 of 27 under 1B on the Decision Index board; both models are on the board. Apache 2.0.

github.comPythonMITpushed oct 10added oct 12

LoRA plus a pointer readout on Qwen3.5 (0.8B / 4B / 9B), Apache 2.0, with training code and frozen eval suites. TypeSafe /v1/systemone drop-in, calibrated by default via a fitted temperature. Frozen evals put Kev-9B about 3.5 points behind Jev on unseen sources; MLX, ROCm, CUDA, and a one-command Modal deploy. Also the only open-weights route on OpenRouter's Decisions API.

github.comPythonApache-2.0pushed oct 12606 forksadded sep 20

0.6B replica trained from scratch, MIT. Ships the weights, the dataset, and the end-to-end training pipeline, plus a side-by-side demo against Jev.

github.comPythonMITpushed sep 21261 forksadded sep 21

Qwen3.5-9B LoRA from Bespoke Labs. Contrastive recipe, not distilled from Jev. 90% vs Jev 93% on a 324-example holdout.

github.comPythonpushed oct 13168 forksadded sep 20

150M ModernBERT encoder that answers per-request noul / choice / score questions in one pass. Trains on a Colab T4 in about 30 minutes. Not the same project as SemIf, which was also once called OpenJev.

github.comPythonApache-2.0pushed oct 62 forksadded sep 20

151M ModernBERT decision engine, non-autoregressive, with published calibration numbers. Claims the top spot over Jev and Laya on the LocalLLaMA typed-decisions benchmark.

github.comPythonpushed sep 2045 forksadded sep 21

Largha-150M: a 144M bilingual decision model on mmBERT-small (English-first, Chinese evaluated on translated cases). Calibrated noul / choice / score answers robust to option reordering, one pass per decision.

github.comPythonApache-2.0pushed oct 13added oct 12

0.8B multimodal Jev-like model that plays GUI games directly from raw pixels. Apache 2.0.

github.comJavaScriptApache-2.0pushed sep 245 forksadded sep 23

Independent reproduction of Jev-style post-training on MiniCPM5-2B via LoRA: 51% → 80.5% on a sealed, leakage-checked holdout for about $5 of compute. Stage 1 covers single-letter choice only; the rigor is the point — significance testing, baselines, and a technical report that voids its own figure after finding contamination. MIT.

github.comPythonMITpushed oct 12added oct 8

Recursively self-improving research loop that trains the models: AI agents propose hypotheses, register predictions before spending GPU time, and retire their own champions on the evidence. Eight releases in 13 days, 496 published experiments including failures; v6.1-VL 27B posts 65.64 on the Decision Index 0.3 public set, the top listed score. MIT code, Apache 2.0 weights, Jev-compatible API.

github.comHTMLMITpushed oct 109 forksadded oct 13

Jev-style replica on Gemma 4 E2B / Gemma 3 270M, trained and served on Modal. One forward pass, no decoding. Live demos for support, Doom, browser-use, and smart home.

github.comPythonpushed sep 176 forksadded sep 20

Together AI's Jev-inspired experiment: Qwen3.5 fine-tunes that take state + question + 2–24 options and return one letter — openly next-token, not a non-autoregressive runtime. MIT, with full data recipe and a "train your own for $17" writeup. A 4B experimental build is also served on OpenRouter.

github.comPythonMITpushed sep 2429 forksadded sep 26

Train-your-own multimodal System One model: evaluates text, images, and video against task instructions and returns probabilities over supplied candidates. Ships the model code, data pipeline, SFT plus experimental RLCD training, and evals.

github.comJupyter NotebookApache-2.0pushed oct 1371 forksadded oct 8

Typed decisions from a frozen LM plus a small physics engine, from Laya's author: calibrated probabilities, conformal answer sets, and an explicit abstain flag. 800M or 4B, 73K context, image inputs.

github.comPythonApache-2.0pushed oct 1125 forksadded oct 12

ModernBERT-base fine-tune with a decision head that answers noul / choice / score and also multi-label questions, with a fitted temperature per question type. Publishes its held-out numbers: 61.2% accuracy and 0.043 ECE on datasets it never trained on. Apache 2.0 code and weights; serves TypeSafe /v1/systemone, so the official SDK works by repointing TYPESAFE_BASE_URL, and an ONNX build runs in the browser.

github.comPythonApache-2.0pushed oct 103 forksadded oct 5

Turns any LLM into a Jev-style decision model: typed decisions with probabilities, no training. From Nokia Applied Research, Apache 2.0.

github.comPythonApache-2.0pushed oct 12152 forksadded sep 23

DiffusionGemma plus vLLM: an inference method rather than new weights, with native image inputs and a hosted API. Ranked 3rd on JevBench v1.2, just under Jev itself.

github.comPythonApache-2.0pushed sep 1911 forksadded sep 22

Self-hosted drop-in on GLiFormer 400M, MIT. Speaks the TypeSafe request shape (JevBench drove it with the official adapter); ranked in v1.2.2.

github.comPythonMITpushed sep 2024 forksadded sep 22

Jev-style parallel constrained decisions for any MLX model on Apple Silicon. Scores every allowed answer per schema field in one pass, so the output is valid by construction.

github.comPythonMITpushed oct 98 forksadded sep 21

Educational Jev-like visual inference experiment on Apple Silicon: shared context, direct candidate scoring, and local visual demos.

github.comPythonMITpushed sep 2129 forksadded sep 23

Adapts local language models into Jev-compatible decision engines with Choice / Score / Noul outputs, via prefill-only binary inference. Apache 2.0.

github.comPythonApache-2.0pushed oct 1145 forksadded sep 23

Ollama for decision models: ollaya pull / serve / run manage Laya, decider, NLI, and GLiClass locally behind a wire-identical TypeSafe /v1/systemone (and /v1/decisions) endpoint.

github.comRustApache-2.0pushed oct 1379 forksadded oct 8

System One-style layer on open weights, ranked on JevBench. Its cautionary detail: 72% vs 21% on the same answer-judging items with option order flipped.

github.comPythonApache-2.0pushed sep 2513 forksadded sep 22

Open, Jev-compatible System One decision server on DiffusionGemma. Apache 2.0.

github.comPythonApache-2.0pushed oct 1361 forksadded sep 23

TypeSafe /v1/systemone on Qwen 35B MoE via SGLang. Prefill once, then first-token logits per question. No extra training.

github.comPythonpushed oct 146 forksadded sep 20

Open recreation on Qwen3.5. Prefills the state once, then scores every question in parallel from next-token logits. Serves the TypeSafe request shape. Browser demo on WebGPU. #3 on Image JevBench in its released stable configuration.

github.comPythonMITpushed oct 618 forksadded sep 20

Runtime-defined semantic decisions from direct option logits. CUDA, MLX, and WebGPU, plus committed benchmarks. Repo recently renamed from SemIf.

github.comPythonMITpushed sep 23353 forksadded sep 24

Turns any open model into a classifier / Jev endpoint. From Featherless, Apache 2.0.

github.comPythonApache-2.0pushed oct 769 forksadded sep 23

Turns any open LLM into a System One classifier: batched single-token choice inference, TypeSafe SDK compatible. Doom and wikiracing demos on Qwen3-8B.

github.comPythonpushed sep 185 forksadded sep 20

Apple's Neural Engine as a free local System One decision engine on macOS: "Jev at home."

github.comPythonpushed sep 22added sep 23

Non-autoregressive local drop-in, sub-15 ms decisions, Apache 2.0.

github.comPythonApache-2.0pushed oct 869 forksadded sep 22