Sovereign Intelligence.
From Software to Hardware — The Complete AI Full Stack.

Eulerwa Inc. opens every layer of AI development — from model training to hardware acceleration to autonomous agents — through CLI and open source.
EulerForge for training, EulerStack for model design, EulerWeave for data, EulerAgent for execution, EulerNPU for hardware, and EulerAtlas for robot learning.

EulerForge EulerStack EulerWeave EulerAgent EulerNPU EulerAtlas

EulerForge: an LLM fine-tuning toolkit that trains dense models in an MoE style

A research-oriented LLM fine-tuning framework that injects LoRA into HuggingFace models and lets you train a dense model as a Mixture-of-LoRAs or MoE Expert LoRA structure. On top of a familiar dense-SFT workflow, EulerForge adds Dense → MoE conversion and phase scheduling, so routing, expert specialization, and MoE stability can be studied reproducibly on a modest GPU budget. A single YAML preset carries you through SFT → DPO/ORPO → RM → PPO.

Learn More Tutorials (24) CLI Reference

Approach — a standardized MoE conversion path for research

EulerForge provides a standardized workflow for converting a dense model into an MoE-style trainable one, so experiments can be expressed in configuration rather than in glue code.

  • Dense → MoE conversion: one YAML switch — mixture_lora / moe_expert_lora
  • LoRA Handoff + Phase scheduling: staged unfreezing (router → LoRA → base FFN) for stable large-model fine-tuning
  • One preset = full pipeline: SFT → DPO/ORPO → RM → PPO in a single command sequence
  • Preflight + MoE stability validation: catches configuration errors before a single GPU cycle burns

4 injection strategies × 5 training paths

Injection Dense LoRA · Mixture LoRA · MoE Expert LoRA · Native MoE Expert LoRA
Training SFT · DPO · ORPO · RM · PPO
Backbones Qwen2/3 · Llama 2/3 · Gemma 3 · Gemma 4 (dense+MoE) · Mixtral
Quantized training nf4 / int4 / int8 (bitsandbytes)
Extras HF Export · Optuna grid/bayesian search · integrated benchmark · 5-language CLI

24 KO/EN tutorials + full CLI reference included

EulerStack: an Architecture Description Language (ADL) for LLMs

EulerStack is an Architecture Description Language for LLMs. It separates architecture out of PyTorch files where structure, training, and serving usually live tangled together — the same abstraction step the semiconductor industry took when Verilog and VHDL replaced schematics-plus-C. A single YAML spec flows through a 5-layer pipeline (DSL → Schema → IR → Compiler → CLI) and lands as a HuggingFace PreTrainedModel (config.json + model.safetensors) that EulerForge can pick up directly for training.

Learn More Tutorials (16) CLI Reference

57 presets on a 3-tier learning path

Organized by the v1 "industrial ordering principle", from validated industrial standards → recent hybrid/MoE → v1 experimental primitives: 24 llm_ (5 sizes × 4–5 variants, MLA included) + 33 arch_ (beginner 2 · intermediate 3 · advanced 5 · expert 23, of which 9 are *_mini). You can experiment with Phase B primitives — MLA, MoD, Titans, Dual-Stream — at arch-scale.

MixersAttention · Mamba · RetNet · Hyena
FFNMLP · Gated MLP (SwiGLU) · MoE (top-k routing)
Skill-level walkthroughbeginner (GPT-2/Llama) · intermediate (Mistral/Gemma2/Qwen) · advanced (Jamba/Samba/RetNet) · expert (MoE × mixer × depth/receptive-field 3D design space)
Compile targetHuggingFace model directory or JSON runtime config

CLI & workflow

# Explore presets
eulerstack presets list
eulerstack presets show llm_2b_simple

# Validate spec (with realism report)
eulerstack validate --preset my_model.yml --report

# Build an HF model dir → train with eulerforge
eulerstack compile --preset my_model.yml --output-dir ./my_model

Three-stage validation (schema → cross-field compatibility → realism heuristics) catches design errors before compilation. All CLI messages are translated into 5 languages (ko/en/zh/ja/es). v0.1.5 adds μP scaling, differentiation auxiliary objectives, and the tissue organ declaration as backward-compatible spec extensions (existing YAML keeps working unchanged).

EulerWeave: The Data Engine

A comprehensive data processing system that bridges the gap between raw datasets and production-ready model training.

Learn More

Open Source: Data Preprocessing

Designed for researchers and developers to refine and analyze data in local environments.

  • Smart Analysis: Initial data profiling using local LLMs.
  • Recursive Filtering: Auto-generation and application of domain-specific filter classes.
  • Metric Computation: PPL, Loss, and MinHash computation for deduplication.
  • Quality Scoring: Noise filtering scores via BERT-based models.

Enterprise: Orchestration

Business-critical features for scaling from a single GPU to cluster-level operations.

  • Scale-Down Validation: Automated 1-GPU LoRA/MoE testing.
  • Resource Estimation: AI-powered GPU-hour and cloud cost calculation.
  • Parameter Tuning: Bayesian optimization and grid search automation.
  • MLOps Integration: End-to-end pipeline from training launch to result synthesis.

EulerAgent: Deterministic CLI Framework

A local-first agentic framework with an 8-state machine, Pattern/Graph orchestrators, RAG, Long-Term Memory (SQLite), MCP integration, 30+ CLI commands, and a plugin system. Every action is logged, audited, and approved by humans. On top of deterministic gate nodes (kind: shell) that run verification as code (not the LLM), it provides a self-repairing coding agent (code.dev_loop.v2) that writes, tests, and fixes its own code.

Learn More

Human-in-the-Loop Security (HITL)

Philosophy: Every action by an agent that changes the world (file writes, shell execution, external calls) must be approved by a human. The Pattern Orchestrator now mixes LLM, judge, and deterministic shell gate nodes (12 canonical patterns); bounded auto-approval (node.auto_approve) and a symbol gate (symbol_check) let it run fix→test loops autonomously.

  • 8-State Machine: CREATE → CONTEXT → PLAN → PROPOSE → APPROVE → EXECUTE → RESULT → FINALIZE.
  • Long-Term Memory (LTM): SQLite-based persistent memory store for user preferences and project context.
  • RAG Pipeline: Hybrid lexical/vector search with citation source tracking.
  • MCP Integration: Model Context Protocol for external tool and search provider connectivity.

Command Line Interface (CLI)

euleragent init
# Create and run an agent
euleragent new my-assistant --template personal-assistant
euleragent run my-assistant --task "Perform code refactoring"

# Audit and approve high-risk actions
euleragent approve list
euleragent approve accept [id] --execute

30+ commands across core, approval, RAG, memory, workflow, pattern, and MCP groups. Built for developers who want autonomous capabilities without sacrificing control and security.

EulerNPU: NPU Inference Composition & Simulation Stack

An NPU inference synthesis stack that composes an operator graph (spec.yaml) from 138 operators across 17 groups, then validates and compiles it (.npuart) to run on a host CPU reference, a functional NPU simulator, and Zynq-7000 (XC7Z020) · Zynq UltraScale+ (ZU3EG/ZU9EG) FPGAs. It provides INT4/INT8 quantization, 10 dtypes, 4 backends, and 15 CLI subcommands. See the product page for a demo of the humanoid-brain SoC (Project 3) being simulated.

Learn More

NPU Operators & Compiler

Supports 138 operators (17 groups, A–Q) and 10 dtypes, compiling spec.yaml into .npuart artifacts.

Operators 138 (Core Math, Activation, Normalization, Conv/Vision, Sequence/Attention, Efficient Attention (Flash/Sliding/GQA), MoE/Sparse, Recurrent, Graph, Multimodal, Vision Encoder, Diffusion, Speculative Decoding, Quantization, Mamba/SSM, Cache Compress, Autonomy)
DType fp32, int32, fp16, bf16, int8, uint8, int16, int4, fp8_e4m3, fp8_e5m2 — INT4 weight quantization supported
Target Zynq-7000 (XC7Z020) · Zynq UltraScale+ (ZU3EG, ZU9EG) FPGA

CLI & Simulation

eulernpu validate spec.yaml
eulernpu compile spec.yaml -o model.npuart
eulernpu quantize model.npuart --weight-bits 4
eulernpu sim model.npuart --cycles
eulernpu profile model.npuart

15 subcommands including validate, compile, run, sim, profile, explain, quantize, migrate-spec, generate, board smoke, calibrate, benchmark, replay, and compress-cache. Backends: cpu_ref · npu_sim · zynq_ps · zynq_pl_stub.

EulerAtlas: Plugin-Based Robot Behavioral Learning (first public: Car · Humanoid)

A plugin-based CLI robot behavioral learning framework running on an RL→IL pipeline that combines imitation learning (IL) with FastTD3 reinforcement learning (RL). The first two flagship domains we are releasing are car (EulerDrive — BEV autonomous driving, CARLA-verified) and humanoid (EulerWalk — RL→IL loco-manipulation), and both ship pretrained models so you can run them with no training at all. The unified schema is designed to cover 8 domains — car, drone, humanoid, robot dog, mobile manipulation, logistics, agriculture, shipyard — and domains keep expanding as plugins.

Learn More

8-Domain Unified Schema

All modules communicate consistently through fixed obs/action dimension schemas per domain.

Classic (01-04) Car 4/2 · Drone 6/3 · Humanoid 33/14 (loco-manip) · Robot Dog 12/8
Industrial (05-08) Mobile Manipulation 18/8 · Warehouse AGV 14/4 · Agri-Robot 16/6 · Shipyard 20/9
Policy Models BC-MLP, BC-RNN, BC-CNN, ACT, Diffusion Policy + SOTA (BEVFuser, TemporalTransformer, WholeBodyACT) + RL Actor
Levels L0 Toy · L1 Intermediate · L2 Advanced (real backends: CARLA/MuJoCo/HumanoidBench)

CLI & Policy Models

euleratlas init --role road --level 0 --out config.yml
euleratlas train -c config.yml
euleratlas train-rl -c config.yml
euleratlas sim rollout -c config.yml --save-video

# Imitation learning + FastTD3 RL · RL→IL · Domain Plugin

Adds train-rl, data collect/generate-demos, and sim rollout --render/--save-video. Includes the Domain Plugin architecture and edge deployment.

Resources & Library

Access specialized technical publications and the open-source ecosystem.

Publications

In-depth research books on quantum computing and AI architectures.

Open Source Ecosystem

Community tools for data processing and model orchestration.

Eulerwa Principles

Technology exists to serve humanity

No Military Automation

We oppose the military automation of artificial intelligence. Technology must be used to protect life, not to take it — no technological achievement can take precedence over human safety.

Humanism & Democracy

Artificial intelligence must serve as a tool that elevates human dignity and strengthens democratic values. We oppose any use of AI that suppresses individual freedom or undermines democratic principles.

Usage Principles

All software, models, and services released by Eulerwa may not be used for purposes that violate these principles. We take our responsibility for how our technology is used seriously.

The Eulerwa Lifecycle

1. EulerStack (Design)
2. EulerForge (Model)
3. EulerWeave (Data)
4. EulerNPU (Inference HW)
5. EulerAtlas (Robot Learning)
6. EulerAgent (Execution)

Build Sovereign AI with Us

Open Source, Local-First, Human-Centric.

Contact Us