Decision Making Python Packages

torchrl

A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.

torchrl-nightly

A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.

emukit

A Python-based toolbox of various methods in decision making, uncertainty quantification and statistical emulation: multi-fidelity, experimental design, Bayesian optimisation, Bayesian quadrature, etc.

futuresearch

Forecast what will happen, including outcomes of your decisions.

solvi

A runtime where decision models propose and your code decides: typed questions, hard rules, quotes checked against the text, escalation with a guaranteed error budget, replayable traces. Works with any System One model.

scikit-decide

AI framework for Reinforcement Learning, Automated Planning and Scheduling

decisio

A serving layer for decisions: typed questions in, a probability for every option out, one forward pass each. Three bases: Qwen3.6-35B-A3B, Gemma 4 12B, Gemma 4 31B.

futuresearch-mcp

Forecast what will happen, including outcomes of your decisions.

flatland-rl

The Flatland Framework is a multi-purpose environment to tackle problems around resilient resource allocation under uncertainty. It is designed to be a flexible and method agnostic to solve a wide range of problems in the field of operations research and reinforcement learning.

laya-cli

Ergonomic CLI for Laya — typed decisions (choice/score/noul) for Humans & AI Agents, with streaming batch mode for pexels-cli pipelines

quantum-reasoning-skill

Model-agnostic Agent Skill: keep multiple hypotheses alive, test them against evidence, revive alternatives, collapse to the best-supported answer. Ships the skill contract, a reference branch controller and a benchmark harness. No model evaluation published yet, so no performance claim.

phocinae-server

Phocinae decision-model inference server: TypeSafe /v1/systemone-compatible local API (Largha 150M)

selfjev

selfjev: an open decisions model (selfjev-4b) with Jev's API. SDK, server, fine-tuning and RLCD.

phocinae-mcp

Phocinae MCP decision server: gate / classify / route / score tools backed by the local Phocinae model

makoto-decision

Decision like hanuma makoto

typesafe-mcp

Typed decision tools for AI agents, powered by TypeSafe Jev

arbitype

Typed decision tools for AI agents, powered by TypeSafe Jev

phocinae-guard

Phocinae command approval gate for AI harnesses (Claude Code / Cline / Codex / Crush / dsh hooks): fail-closed local decision guard

abtree

Asynchronous behavior tree framework with behavior forest collaboration,built on Python asyncio for multi-agent systems and autonomous decision-making.

decguard

Reliability testing for probabilistic AI decisions in system-one-models.

arbitration-graphs

Hierarchical behavior models for complex decision-making and behavior generation in robotics

eq-layer

A control layer that decides how an LLM should speak before it speaks

ink-jit

Behavior JIT for production AI. Turn verified repeated decisions into qualified local software.

parley-consensus

Provable consensus for AI agents of rival owners: deterministic red lines, max-min consensus over masked verdicts, and a verifiable non-betrayal transcript. Multilateral, non-crypto, self-hosted.