--- title: JEV-Gemma4-26B-A4B canonical_url: "https://www.modelscope.ai/models/AutoTrust-Labs/JEV-Gemma4-26B-A4B" md_url: "https://www.modelscope.ai/models/AutoTrust-Labs/JEV-Gemma4-26B-A4B.md" repository: AutoTrust-Labs/JEV-Gemma4-26B-A4B last_updated: 2026-10-12 license: apache-2.0 pipeline_tag: text-classification tasks: - text-classification model_type: - gemma4 architectures: - Gemma4ForConditionalGeneration base_model: - google/gemma-4-26B-A4B-it base_model_relation: adapter parameters: 25.8B tensor_type: - BF16 library_name: - transformer - lora - safetensors - pytorch frameworks: - pytorch language: - en downloads: 37 stars: 0 tags: - system-one - system-two - typed-decisions - decision-model - calibrated-probabilities - jev - noul - choice - score - lora - gemma4 - mixture-of-experts --- # JEV-Gemma4-26B-A4B > JEV-Gemma4-26B-A4B - An open-source model by AutoTrust-Labs on ModelScope. autotrust/JEV-Gemma4-26B-A4B AutoTrust-Labs/JEV-Gemma4-26B-A4B is a 25.8B-parameter text-classification model on ModelScope. licensed under apache-2.0. derived from google/gemma-4-26B-A4B-it. - **Repository**: AutoTrust-Labs/JEV-Gemma4-26B-A4B - **License**: apache-2.0 - **Tasks**: text-classification - **Parameters**: 25.8B - **Base model**: google/gemma-4-26B-A4B-it - **Tags**: system-one, system-two, typed-decisions, decision-model, calibrated-probabilities, jev, noul, choice, score, lora, gemma4, mixture-of-experts - **Downloads**: 37 - **Stars**: 0 - **Last updated**: 2026-10-12 Source: https://www.modelscope.ai/models/AutoTrust-Labs/JEV-Gemma4-26B-A4B --- # autotrust/JEV-Gemma4-26B-A4B [![AutoTrust/GuruSearch recommendation & search live demo: news.guru.so](https://huggingface.co/autotrust/JEV-27B/resolve/main/assets/autotrust_gurusearch_demo.png)](https://news.guru.so/ui) ## 🔴 Live: AutoTrust/GuruSearch recommendation & search demo → [news.guru.so](https://news.guru.so/ui) > [!IMPORTANT] > **Try AutoTrust/GuruSearch live (new, 11 October 2026).** A recommendation and search demo in which every ranking is a > calibrated System 1 decision, made in about half a second: > * **Recommendation:** the latest headlines from 6 news feeds, ranked by importance. > * **Search:** results from Google News, Bing News and Yahoo News re-ranked by relevance, side by side with the > search engines' own order, plus a short answer with citations. ### An integrated System 1 + System 2 open model on Gemma-4-26B-A4B-it — Decision Index 0.2.1: **58.05** This bundle serves two systems from one set of base weights. **System 2** is the unmodified `google/gemma-4-26B-A4B-it` (a 26 B-parameter mixture-of-experts model with ≈ 4 B active parameters per token, text and image input): generation and reasoning, bit-identical to the base model. **System 1** adds a LoRA adapter and a 24-slot decision head and returns calibrated probabilities for three typed primitives in one prefill pass: * `noul`: true / false * `choice`: 2–16 options (wider questions are read in groups of ≤ 16 plus a final) * `score`: 0–5 > **Two models, two organisations.** **TypeSafe Jev 1.13** is the hosted, closed model made by TypeSafe AI. > **autotrust/JEV-Gemma4-26B-A4B** is an independent open-weights model built by AutoTrust AI; it is not affiliated > with, endorsed by, or a product of TypeSafe AI. ## Decision Index 0.2.1 Complete run of all 150,759 requests of suite 0.2 (150,317 scored; 0 errors, 0 unsupported), scored with the kit's `score --edition 0.2.1`. Results: [`autotrust/jev-decision-index-results`](https://huggingface.co/datasets/autotrust/jev-decision-index-results) (`runs/jev-gemma4-26b-a4b`). | | Decision Index (balanced skill) | balanced raw | breadth skill | |---|---:|---:|---:| | **autotrust/JEV-Gemma4-26B-A4B** | **58.05** | 67.35 | 56.98 | | TypeSafe Jev 1.13 (board) | 57.91 | — | — | | [autotrust/JEV-27B](https://huggingface.co/autotrust/JEV-27B) | 53.30 | 64.32 | 52.21 | | area (skill) | Knowledge & Reasoning | Language | Retrieval & Classification | Tools & Automation | Arts & Taste | |---|---:|---:|---:|---:|---:| | JEV-Gemma4-26B-A4B | 0.430 | 0.636 | 0.679 | 0.697 | 0.415 | ## Engine and read-out * **Template** `bare-v1`, prefixed with ``: `[kind] … [state] … [question] … [options] A) … [decision]:` * **Read-out**: the final-norm hidden state of the last token goes through a linear fp32 head (hidden 2,816 → 24 slots), soft-capped at 30 like Gemma's own logits; inactive slots are masked, the logits are divided by the per-kind temperature and softmaxed. Nothing is generated. * **Adapter**: `adapter/` is merged into the bf16 backbone in memory at load (as in the Decision Index run). * **More than 16 options**: ⌈n/16⌉ contiguous groups read whole, then a final of 16; every option is read, none pruned. The reference engine for the Decision Index is `submissions/jev/jev_engine.py` in the [Decision Index kit](https://github.com/apolinario/decision-index). ## Temperatures | table | noul | choice | score | |---|---|---|---| | `calibration.json` (default; used in the Decision Index run) | 1.003 | 1.017 | 0.999 | | `calibration_gold.json` (calibrated against ground-truth answers) | 1.214 | 1.098 | 1.000 | Use `calibration_gold.json` when you gate automatic actions on confidence. ## Usage (transformers + peft) ```python import json, torch from huggingface_hub import snapshot_download from peft import PeftModel from safetensors.torch import load_file from transformers import AutoTokenizer, Gemma4ForConditionalGeneration d = snapshot_download("autotrust/JEV-Gemma4-26B-A4B") tok = AutoTokenizer.from_pretrained(d) base = Gemma4ForConditionalGeneration.from_pretrained(d, dtype=torch.bfloat16, device_map="cuda") # System 2: `base` is gemma-4-26B-A4B-it unchanged — use base.generate(...) (text or images). # System 1: adapter merged in memory + head m = PeftModel.from_pretrained(base, f"{d}/adapter").merge_and_unload().eval() backbone = m.model jc, T = json.load(open(f"{d}/judge_config.json")), json.load(open(f"{d}/calibration.json"))["per_kind"] head = load_file(f"{d}/head.safetensors"); W, b = head["proj.weight"].cuda(), head["proj.bias"].cuda() @torch.no_grad() def decide(kind, state, question, options): lines = options if kind != "choice" else [f"{'ABCDEFGHIJKLMNOP'[i]}) {o}" for i, o in enumerate(options)] text = f"[kind] {kind}\n[state] {state}\n[question] {question}\n[options]\n" + "\n".join(lines) + "\n[decision]:" ids = torch.tensor([[tok.bos_token_id] + tok.encode(text, add_special_tokens=False)], device="cuda") h = backbone(input_ids=ids, use_cache=False).last_hidden_state[0, -1].float() z = 30.0 * torch.tanh((W @ h + b) / 30.0) s, _ = jc["slots"]["ranges"][kind] return dict(zip(options, torch.softmax(z[s:s + len(options)] / T[kind], 0).tolist())) print(decide("noul", "Customer says the parcel arrived damaged and wants their money back.", "Is the customer asking for a refund?", ["false", "true"])) ``` ## Training data (disclosure) System 1 was trained on teacher distributions and ground-truth decision data. The ground-truth data includes the public **training splits** of some datasets whose **test** splits the Decision Index uses; no test split of any benchmark was used, and suite items were excluded before training. The list has been provided to the Decision Index maintainers with the submission. MMMU / MMMU-Pro are not valid evaluations for this model. ## Limitations * Where the teacher is wrong, System 1 often is too. * Weaker than JEV-27B on long structured inputs (e.g. the Decision Index's Home appliance simulator and POP909). * HLE: below chance, like every open entry on the board; do not use System 1 for expert-level questions. * `noul` and `score` accept only their canonical options. * English-centric; not for high-stakes decisions without confidence gating. ## Files ``` model-*.safetensors · config.json · processor_config.json · tokenizer* · chat_template.jinja google/gemma-4-26B-A4B-it, unchanged (System 2; text + image input) adapter/ System 1 LoRA (peft) head.safetensors 24-slot decision head (fp32): proj.weight [24, 2816], proj.bias [24] judge_config.json slot layout, verbalizer ids, softcap, read-out, provenance calibration.json per-kind temperatures (default) calibration_gold.json per-kind temperatures calibrated against ground-truth answers reports/ Decision Index scores ``` ## License Apache-2.0 for the adapter, head and calibration files; base model under the Gemma 4 terms (). Not affiliated with TypeSafe AI.