--- license: apache-2.0 language: - en - pt library_name: transformers pipeline_tag: text-classification base_model: Tropic-AI/moBERTo-orig-tokenizer tags: - typestrict - decision-model - typed-decisions - encoder - decision-index - onnx - cpu ---

TS-moBERTo-150M by typestrict.ai

Train your own on typestrict.ai | Docs | MCP for AI assistants | CLI | All TS models
License: Apache 2.0 | Authors: typestrict.ai

# TS-moBERTo-150M Our default base: one encoder that reads a question, a state and a set of options and returns a calibrated probability for every option, in English and Portuguese, fast enough for a CPU. It was published before as Dinah-1. > [!TIP] > **Train on it today** at [typestrict.ai](https://typestrict.ai): it is the default base of every new training. - **Typed decisions**: `choice` (one option out of many), `noul` (is this statement true) and `score` (a value on an ordered rubric), each with the probability of every option and a confidence. - **Small and fast**: 150M parameters, English and Portuguese, 8,192 tokens of context, with ONNX files that run on a CPU. - **Open**: Apache-2.0 weights, trained only on data whose license allows it (listed below). - **The base of every typestrict model**: send your data to [typestrict.ai](https://typestrict.ai) and get a model that answers your own question, measured against Jev on your held-out examples.

Decision Index 0.3 of the TS models

## 🧭 The TS family Every TS model follows the same recipe on the same data, so trying another one is a new version of the same training.
ModelSizeLanguagesOn typestrict.ai
TS-moBERTo-150M150MEnglish and PortugueseAvailable, recommended
TS-Ettin-150M150MEnglishAvailable
TS-Ettin-68M68MEnglishAvailable
TS-Ettin-32M32MEnglishAvailable
TS-Ettin-17M17MEnglishAvailable
## 📊 Results ### Decision Index 0.3 The public index of the [Decision Index](https://github.com/apolinario/decision-index) 0.3, in our lab's run: the encoder alone, then with the code and calculator experts that ship with [TS-moBERTo-150M](https://huggingface.co/typestrict/TS-moBERTo-150M). Measured on public benchmarks, not on your task: a model trained on typestrict is measured on your own examples.
ModelEncoder aloneWith code and calculator experts
TS-moBERTo-150M31.2033.89
TS-Ettin-150M31.9634.78
TS-Ettin-68M27.6130.46
TS-Ettin-32M19.7022.45
TS-Ettin-17M15.7618.52
The official run of the Decision Index kit on this model, published as [Dinah-1](https://huggingface.co/Lukitaduarte/dinah-1), scored **34.32**, with the raw results open for auditing in [Lukitaduarte/dinah-1-decision-index-results](https://huggingface.co/datasets/Lukitaduarte/dinah-1-decision-index-results). ### Held-out typed decisions Held-out sets of 1,000 items each, accuracy:
SetAccuracy
Typed decisions, English0.790
Typed decisions, Portuguese0.780
Chess puzzles0.585
## 🏃 Quickstart The inference code ships in this repository (`dinah.py`). On a CPU, with the int8 ONNX file: ```bash pip install huggingface_hub onnxruntime transformers numpy ``` ```python import importlib.util from huggingface_hub import hf_hub_download REPO = "typestrict/TS-moBERTo-150M" spec = importlib.util.spec_from_file_location("dinah", hf_hub_download(REPO, "dinah.py")) dinah = importlib.util.module_from_spec(spec) spec.loader.exec_module(dinah) model = dinah.DinahONNX.from_pretrained(REPO) # or dinah.Dinah for PyTorch print(model.choice( state="My card still has not arrived. How can I track the delivery?", criteria={ "payment": "A problem with a payment or a charge", "card_delivery": "Delivery or shipping of the card", "cancel": "Cancel the card or the account", }, instructions="What is the customer's message about?", )) # What this model answers here: "card_delivery", probability 0.999, confidence 0.979 ``` `noul` and `score` work the same way (`model.noul(...)`, `model.score(...)`); `model.predict([...])` answers a batch. This is the base model, before any training on your data: for your own decision, train it on [typestrict.ai](https://typestrict.ai). ## 🗂️ History - Published on 2026-10-14 as [Dinah-1](https://huggingface.co/Lukitaduarte/dinah-1) (revision `b518a5e7`), renamed TS-moBERTo-150M on 2026-10-19. - The weights are identical: the same SHA-256 for the encoder, the calculator expert and both ONNX files. Nothing was retrained. - The results published for Dinah-1, such as the official Decision Index run (34.32), apply to this model. - Use `typestrict/TS-moBERTo-150M` in new projects; the Dinah-1 repository stays online for citations and existing integrations. ## 🚀 Train your own on typestrict.ai

Train a decision model on your data with typestrict.ai

TS-moBERTo-150M is a base: it already reads language and makes typed decisions. On [typestrict.ai](https://typestrict.ai) it learns **your** decision (your queues, teams, products, intents) from your data, and the result is measured against Jev, the market reference model, on examples held out from training. If it does not beat Jev, the credit comes back. **In the web app**: upload a spreadsheet, a folder of documents or a public dataset at [typestrict.ai](https://typestrict.ai), review what the agent proposes, and train. **From your AI assistant** (ChatGPT, Claude, Cursor, Codex, Copilot and others), with the remote MCP server: ```bash claude mcp add --transport http typestrict https://api.typestrict.ai/mcp ``` **From a terminal or CI**, with the [typestrict CLI](https://github.com/typestrict-lab/cli): ```bash curl -fsSL https://typestrict.ai/install.sh | sh && typestrict login typestrict train tickets.csv --goal "Which team handles the ticket?" --base ts-moberto-150m --watch ``` The trained model comes as a small package (`decision_ai.json`, `tokenizer.json`, `onnx/model_8bit.onnx`) that runs on your own servers, in a container, or inside a Flutter app with [decision_ai](https://pub.dev/packages/decision_ai). Setup for every assistant: [docs.typestrict.ai/agents](https://docs.typestrict.ai/agents). ## 🗒️ Model details
PropertyTS-moBERTo-150M
Parameters150M
ArchitectureBidirectional encoder with a typed-decision head
Starts frommoBERTo (Apache-2.0)
LanguagesEnglish and Portuguese
Context length8,192 tokens (longer examples are refused, never cut)
Question typeschoice, noul, score, with probabilities and a confidence
Runs onCPU (ONNX) or GPU (PyTorch)
typestrict idts-moberto-150m
LicenseApache-2.0
## 📚 Training data Two stages (2,048 tokens, then 8,192) and an average of both, on the same mixture for every TS model. Each source with the license its own card declares; sources under CC BY-SA and CDLA-Sharing are used for training only and are not redistributed. The training blocks were checked against the Decision Index's public test rows (exact and near-duplicate removal, and 8-gram overlap for the newest blocks).
AreaSources
Typed decisionsZefanCai/Open-Jev (CC0-1.0), LocalLLaMA/typed-decisions (Apache-2.0), typed-decisions-pt translated with Qwen3-4B (Apache-2.0), tasksource/procedural-typed-decisions (Apache-2.0)
Intents and customer supportAmazonScience/massive (CC BY 4.0), Bitext customer support (CDLA-Sharing-1.0), BANKING77 (CC BY 4.0), CLINC150 (CC BY 3.0), FAQ_BACEN (Apache-2.0)
Inference and fact checkingWANLI (CC BY 4.0), SNLI (CC BY-SA 4.0), HoVer (CC BY-SA 4.0), FEVER-NLI (MIT; FEVER and Wikipedia CC BY-SA 3.0), ContractNLI (CC BY 4.0), bAbI-NLI (BSD), FaQuAD-NLI (CC BY 4.0), ExtraGLUE RTE and COPA in Portuguese (MIT)
Reasoning and knowledgeCommonsenseQA (MIT), Cosmos QA (CC BY 4.0), BoolQ (CC BY-SA 3.0), HellaSwag (MIT), WinoGrande (Apache-2.0), ARC train split (CC BY-SA 4.0), QASC (CC BY 4.0), StrategyQA (MIT), BIG-bench tasks without BBH (Apache-2.0), MMLU auxiliary train, ARC and OpenBookQA parts (MIT), GSM8K (MIT), CLadder (MIT)
Health, finance and lawPubMedQA (MIT), MedMCQA (Apache-2.0), twitter-financial-news-sentiment (MIT), financial-tweets-sentiment (MIT), NOSIBLE financial-sentiment (ODC-By), FinQA hallucination detection (MIT), pyRis/SEntFiN (MIT), CUAD (CC BY 4.0), jurisprudencias_br (CC BY 4.0), IBM Claim Stance (CC BY 3.0)
Sentiment, emotion and moderationGoEmotions (Apache-2.0), Civil Comments (CC0-1.0), iSarcasmEval (MIT), B2W-Reviews01 (CC BY 4.0), multilingual-sentiments in Portuguese (Apache-2.0)
Retrieval and hallucinationESCI (Apache-2.0), HaluEval (Apache-2.0), RAGTruth (MIT), CNN/DailyMail (Apache-2.0), SQuAD 2.0 (CC BY-SA 4.0)
Toolsglaive-function-calling-v2 (Apache-2.0), ToolACE (Apache-2.0), hermes-function-calling-v1 (Apache-2.0), nvidia/When2Call (CC BY 4.0)
PhishingLLMGen-Phishing-Email-Dataset (Apache-2.0), aura-phishing-email-corpus, SpamAssassin part (CC BY 4.0)
ChessLichess puzzles (CC0-1.0), google-deepmind/searchless_chess (Apache-2.0)
Humornewyorker_caption_contest, matching task (CC BY 4.0)
Made by usEnglish to Portuguese translations with opus-mt-tc-big-en-pt (CC BY 4.0); synthetic phishing emails written with gpt-oss-120b, labels set by code (Apache-2.0); synthetic home-appliance households, RAG hallucination examples and programmatic skills (arithmetic, dates, units, sorting, graphs, code output), generated by our scripts (MIT and Apache-2.0)
## Links - **typestrict.ai**: [typestrict.ai](https://typestrict.ai), the docs at [docs.typestrict.ai](https://docs.typestrict.ai) (for assistants: [llms-full.txt](https://docs.typestrict.ai/llms-full.txt)) - **Open source**: [CLI and MCP server](https://github.com/typestrict-lab/cli), [skills for AI assistants](https://github.com/typestrict-lab/skills), [decision_ai for Flutter](https://pub.dev/packages/decision_ai) - **Contact**: [typestrict.ai/contact](https://typestrict.ai/contact) ## Citation ```bibtex @misc{typestrict2026tsmoberto150m, author = {{typestrict.ai}}, title = {TS-moBERTo-150M: a typed-decision encoder}, year = {2026}, url = {https://huggingface.co/typestrict/TS-moBERTo-150M} } ```