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: 0.4.10
Nandakishora4958de6ad4f
diff --git a/README.md b/README.mdindex c099b370..f33c9cec 100644--- a/README.md+++ b/README.md@@ -94,33 +94,21 @@ 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 0.4.9--* **A standalone Python HTTP client.** `sdk/python` holds `laya-client`: synchronous `predict()`- and `health()` against a self-hosted `laya-serve`, written on the standard library with zero- dependencies, so an application server no longer installs torch to call one (#1264, issue- #1176). It refuses redirects so a bearer token is never forwarded, rejects a `base_url`- carrying credentials or whitespace, and blocks non-finite JSON in both directions. Install it- from this checkout with `python -m pip install ./sdk/python`; it is not on PyPI yet.-* **`laya-go` gains the confidence gate, hooks and digest verification.** `WithMinConfidence`- applies the same abstention gate as the Python runtime, `OnPredictStart`/`End`/`Error` dispatch- with timeouts and panic isolation, per-language temperatures are read from the checkpoint, and- `laya-go/internal/digest` checks a checkpoint's SHA-256 before loading it, streamed, with path- traversal refused (#1247, #1260).-* **`LAYA_DEVICE=cuda:258` is refused rather than silently asking for GPU 2.** torch keeps a CUDA- index in its low 8 bits, so it parses that string and builds `cuda:2`, and `cuda:128` builds- `cuda:-128`. A value whose parsed device is not the device it names now raises, naming the- variable (#1242).-* **Three more gate parameters that could not fail.** A tolerance of `Infinity` accepts a drift of- any size while the run still exits 0 (#1250), `--revision` naming two different commits for one- checkpoint is refused instead of resolved to whichever came last (#1248), and a slice label- containing `|` is escaped so a tag like `risk|high` cannot shift every number in its row of a- `--markdown` report (#1241). A `Dataset` row built in Python now gets the checks a JSONL row- already had, before a single inference call rather than after (#1253).-* **`LAYA_CPU_AMP=bf16` is documented as a hardware requirement, not a preference.** On a CPU- without native BF16 instructions the forward dies with an illegal instruction rather than- raising, so `_infer`'s autocast retry cannot catch it. `docs/reference/agent.md` now carries the- capability reading to check and a child-process probe to qualify a host.+## What's new in 0.4.10++* **A clamped temperature fit says so instead of looking like a finished fit.**+ `fit_temperature_map` now returns `temperature_clamped`, naming every bucket or question type+ whose optimum fell outside `[0.5, 5.0]`, with the unclamped value beside the one installed. The+ clamped value is still what to install, since checkpoint load applies the same bounds, but a+ fitted 0.387 arriving as exactly 0.5 was indistinguishable from a fit that really wanted 0.5, so+ anyone calibrating their own checkpoint could not tell the floor from an answer (issue #1245).+* **The `labels` guidance in `docs/questions-and-answers.md` was backwards for one real case.** A+ `noul` instruction asking whether something is "positive" answers no for every state at+ confidence 1.0 on both PyTorch checkpoints, and naming the two options through `labels` is the+ one thing that restores it; rewording does not reliably, and capitalising the word helps in some+ wordings and not others. The doc said labels are not a way to fix a question the model gets+ wrong, which is still true in the other direction, and both directions are now stated (issue+ #1244).Earlier releases are in the [GitHub releases](https://github.com/NandhaKishorM/laya/releases).diff --git a/laya/__init__.py b/laya/__init__.pyindex 4488cb75..43690a98 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__ = "0.4.9"+__version__ = "0.4.10"# 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 b5fee1c3..56b09a5e 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 = "0.4.9";+ version = "0.4.10";src = ../.;format = "setuptools";propagatedBuildInputs = with python3Packages; [diff --git a/pyproject.toml b/pyproject.tomlindex 1c358b68..6c0d0e61 100644--- a/pyproject.toml+++ b/pyproject.toml@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"[project]name = "laya"-version = "0.4.9"+version = "0.4.10"description = "Fast, non-autoregressive System 1 decision engine with calibrated probabilities"readme = "README.md"requires-python = ">=3.10"