Open-Weight Models

Model

jev-qwen

Patrick Dugan

LoRA adapters from the Small Jev experiments (Patrick Dugan, Red Team Gladiatorics). TypeSafe's Jev is described as a "System One" model: it returns a typed, calibrated decision over a pre-enumerated output space from a single query. These adapters give an open 4B base model a single-pass, option-scored readout of that kind. They are benchmarked against compact looped decoders, TRM skill gates, script/LDT/Control-Harness gates and a 1-bit 8B LLM. This is not Jev. Jev's weights and architecture are not public; nothing here measures Jev. published, because they contain held-out evaluation items of other projects. All adapters are LoRA r16 / alpha 32 on q/k/v/o, the DeltaNet projections…

A System One decision model: it does not write text. It reads a state, answers typed questions (choice, score, noul), and returns calibrated probability distributions your code can branch on. This repo is a rank-16 LoRA (34,881,536 parameters) on google/gemma-4-E4B-it, merged into the weights at load. How it was trained (readout fine-tuning). The model is trained on its own decision readout — the distribution over the allowed answers read at the answer position, one forward pass, no decoding — with the primitive's proper scoring rule, plus a coherence penalty (weight 1.0): every training question comes with automatically derived siblings (the options as yes/no questions, the negation, the…

A Jevify Tier 2 model: Qwen/Qwen3-VL-2B-Instruct used as a vision-language model: state may carry images (PIL, path, URL, data URI or bytes) and the same three typed questions are asked about them, with a rank-16 LoRA confined to the decoder half of the model, trained on the readout itself — the restricted log-softmax over the allowed answers at the answer position — with the primitive's proper scoring rule, on the aokvqa train split only. The adapter (17,432,576 parameters) is merged into the backbone at load, so it serves at the plain checkpoint's speed. The calibration recipe was fitted on validation splits of the adapted model. pope, ai2d never appeared in training. Serve it as a…

A System One decision model: it does not write text. It reads a state, answers typed questions (choice, score, noul), and returns calibrated probability distributions your code can branch on. This repo is a rank-16 LoRA (21,233,664 parameters) on Qwen/Qwen3.5-4B-Base, merged into the weights at load. How it was trained (readout fine-tuning). The model is trained on its own decision readout — the distribution over the allowed answers read at the answer position, one forward pass, no decoding — with the primitive's proper scoring rule, plus a coherence penalty (weight 1.0): every training question comes with automatically derived siblings (the options as yes/no questions, the negation, the…

A System One decision model: it does not write text. It reads a state, answers typed questions (choice, score, noul), and returns calibrated probability distributions your code can branch on. This repo is a rank-16 LoRA (21,233,664 parameters) on Qwen/Qwen3.5-4B, merged into the weights at load. How it was trained (readout fine-tuning). The model is trained on its own decision readout — the distribution over the allowed answers read at the answer position, one forward pass, no decoding — with the primitive's proper scoring rule. Options are shuffled per family. Training data: the train splits of the 16 non-held-out jev-bench sources (5,885 families, at most 400 records per source); lr…

A System One decision model: it does not write text. It reads a state, answers typed questions (choice, score, noul), and returns calibrated probability distributions your code can branch on. This repo is a rank-16 LoRA (21,233,664 parameters) on Qwen/Qwen3.5-4B, merged into the weights at load. How it was trained (readout fine-tuning). The model is trained on its own decision readout — the distribution over the allowed answers read at the answer position, one forward pass, no decoding — with the primitive's proper scoring rule, plus a coherence penalty (weight 1.0): every training question comes with automatically derived siblings (the options as yes/no questions, the negation, the…

ONNX conversion of jinaai/jina-code-embeddings-0.5b (Qwen2.5-Coder-0.5B finetune, 494M params), laid out like weights under onnx/, tokenizer + config at root, transformers.js compatible. Dynamic int8 quantization was tried (ORT quantizedynamic, QInt8, per-channel + reduce-range, with and without quantpreprocess shape inference) and scored drift ≈ 0.145 — two orders of magnitude over the 1e-3 bar. This GQA-decoder + RMSNorm + last-token-pooling architecture is far more quant-sensitive than BERT-style encoders where int8 dynamic works. Per the sentence-transformers team's guidance, the recommended efficiency path for embeddings is embedding quantization (quantize output vectors to int8/binary…

Model · Text classification

jiwo-0.8b-GGUF

Breaksoftware

I decided to use jiwo-0.8b because it had a high score on https://huggingface.co/spaces/multimodalart/jev-decision-index compared to its parameter count. This GGUF conversion / quantization was generated by a local agent powered by Qwen3.8 27B. See agent-trajectory-log.md for exhaustive details. Run with https://llama.app This is a decision model, to be used via /v1/systemone API. See https://github.com/ggml-org/llama.cpp/pull/29818 - https://huggingface.co/eljiwo/jiwo-0.8b

Model

jojo-sovereign-strong

Mohammed Abdullah H Abdulaziz

This model was finetuned and converted to GGUF format using Unsloth. - Meta-Llama-3.1-8B-Instruct.Q4KM.gguf An Ollama Modelfile is included for easy deployment. The model's BOS token behavior was adjusted for GGUF compatibility. This was trained 2x faster with Unsloth

This model is a fine-tuned version of francesca9805/ppt-wc-zipf-newlex-jpn-after-100mb-packed-bfdisoseed455. It has been trained using TRL. This model was trained with SFT.

Artifacts from round 1 of J-space character distillation on google/gemma-3-12b-it (rev 96b6f1ec). Code, configs, run log and deviations: https://github.com/arav-dhoot/introspection-training/tree/jspace-distillation-arav/jspace-dist (commit 1bf6a9b); see REPRODUCE.md there. Layout mirrors the repo's results/ directory: - teacherseed0/adapters/final: LoRA teachers. TEM = teacheremfinanciallr5e-5seed0 (risky financial advice, 37.2% misaligned), control = teacherctrllr5e-5seed0. - students{1..5}seed0/adapters/{ckpt1000,ckpt5000,ckpt10000,final}: main students. - lrsweepseed0/, vseed0/ (readout-space variants, ckpt 250/500/1000). - gatee/gatee/baserefit/lens.pt (J-lens refit of the base model)…

Experimental multilingual fine-tune of Julia-1 for fast bounded-decision routing on AMD Strix Halo-class local machines. This is an experimental v0.2 multilingual candidate, not a replacement for the English-focused v0.1 checkpoint. - task routing over fixed options - documentation-update triage - context-management metadata - non-authoritative tool-policy hints Do not use this model as the final authority for destructive commands, credential handling, production deploys, security replay safety, durable memory writes, summarization, or final prose. By language on the large multilingual holdout: All latency numbers in development were measured on CPU. NPU acceleration has not been validated.…

Final-iteration draft weights for six single-variable arms trained against moonshotai/Kimi-K3-Flash on the GB300 cluster. Staged for batch export + config, and the draft architecture JSON. optimizer/ state is deliberately excluded — it is ~80% of the bytes and only needed to resume training. from its parent by one thing, so adjacent arms are differenceable. g3j is the one arm that varies initialisation rather than architecture: it has g3f's draft JSON byte-for-byte but starts from g3e's converged weights instead of scratch, so g3f vs g3j isolates warm-start. Note it loads block-5 weights and then trains at block 3 — the load was clean (no shape warnings in the trainer log), but serve it…

ONNX int8 export of kyutai/pocket-tts, six bundles (german2026-10, english2026-09, french2026-10, italian2026-10, spanish2026-10, portuguese2026-10). Source: kyutai/pocket-tts @ 3e82814a68665eec246ff649b14c71331f955c06 (languages/ /model.safetensors). The model weights are © Kyutai Labs, licensed under CC BY 4.0 (inherited from kyutai/pocket-tts). The ONNX conversion and the quantization are the only changes. Please credit Kyutai Labs when you use these files.