Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.
Python calibrationclassificationdecision-modelhuggingfacejevmodernbertmultilingualnlppythonpytorchroutingtyped-decisionszero-shot
release: 0.4.1
Nandakishor1adc59f7e371
diff --git a/README.md b/README.mdindex 7023e368..0b3ae358 100644--- a/README.md+++ b/README.md@@ -94,15 +94,18 @@ The shipped checkpoints work zero-shot, but fine-tuning on decisions from your o**[nandhakishorm.github.io/laya](https://nandhakishorm.github.io/laya/)**: guides for [prediction hooks](https://nandhakishorm.github.io/laya/hooks/), [schema-driven decisions](https://nandhakishorm.github.io/laya/structured/), [Docker](https://nandhakishorm.github.io/laya/docker/) and [LangChain and LangGraph](https://nandhakishorm.github.io/laya/langchain/), plus a full [API reference](https://nandhakishorm.github.io/laya/reference/).-## What's new in 0.4.0--* **The routing default is now `multilingual`.** Text whose language the detector cannot place- used to fall back to the English checkpoint. On the 51-language sweep the multilingual- checkpoint leads on 50 of the 51, the exception being English itself (0.820 against 0.710),- and by 0.180 macro accuracy excluding English. Identified English still routes to the English- checkpoint, so only undecided text moves. Set `Router(default="english")` or- `LAYA_DEFAULT_MODEL=english` to restore the old behaviour; the break-even is around 62%- English traffic.+## What's new in 0.4.1++* **Serve your own checkpoints.** `LAYA_EXTRA_MODELS` registers extra checkpoints on+ `laya-serve` as a JSON name-to-source map, and a registered name pins its checkpoint over+ HTTP like a built-in (#1047, #919). Malformed values stop the server at startup.+* **Fine-tuning tells you its budget.** A run now prints its optimizer-update count before it+ starts: 5,600 items gives 352 updates where 1,100 gives 72 at the same settings, which is why+ small datasets collapsed silently (#1014, #963).+* **Fixes.** `laya-evals` writes redirected output as utf-8 (#1045); the server serves again+ after a lifespan restart (#1049); `Router.decide_batch` takes a checkpoint pin (#1050); the+ ONNX path answers single-option questions instead of failing (#1051).+* **Docs.** Twenty-four corrections where a page contradicted the code, each verified against it.Earlier releases are in the [GitHub releases](https://github.com/NandhaKishorM/laya/releases).diff --git a/laya/__init__.py b/laya/__init__.pyindex f42f5bd0..16e4ef2e 100644--- a/laya/__init__.py+++ b/laya/__init__.py@@ -35,7 +35,7 @@ from .presets import (from .router import DEFAULT_MODELS, RouteDecision, Routerfrom .structured import DecisionResult, decide, decide_batch-__version__ = "0.4.0"+__version__ = "0.4.1"# Routing, language detection and email cleaning are pure Python. The torch-backed names are# resolved lazily so that `import laya` -- and therefore `from laya import Router` ordiff --git a/nix/package.nix b/nix/package.nixindex f18512fc..4960c758 100644--- a/nix/package.nix+++ b/nix/package.nix@@ -12,7 +12,7 @@ let# side, where tests/test_packaging.py compares all three declarations (this one,# pyproject.toml and laya.__version__) and fails CI when they diverge. Nothing# did, which is how this sat at 0.3.4 while the package reached 0.3.20.- version = "0.4.0";+ version = "0.4.1";src = ../.;format = "setuptools";propagatedBuildInputs = with python3Packages; [diff --git a/pyproject.toml b/pyproject.tomlindex 71ac3df7..0dc284b9 100644--- a/pyproject.toml+++ b/pyproject.toml@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"[project]name = "laya"-version = "0.4.0"+version = "0.4.1"description = "Fast, non-autoregressive System 1 decision engine with calibrated probabilities"readme = "README.md"requires-python = ">=3.10"