Onyx Forge | AI Product Factory
Build, train, evaluate, and deploy vertical AI products. From domain expertise to production-ready deployment packages — with cryptographic receipts at every stage. Runs locally on your hardware.
Data Ingest view — Forge running locally on Mac Studio, parsing documents into RAG-ready embeddings
Full lifecycle, one dashboard
Eleven modules covering every stage of AI product development — from data ingestion to production monitoring. Each module produces verifiable artifacts.
From expertise to deployment
Six stages. Each produces verifiable artifacts. Every artifact is cryptographically signed. No black boxes — just auditable, reproducible AI product engineering.
Data Ingest
01Parse, chunk, embed, and store documents into the vector database. Supports PDF, DOCX, HTML, JSON, CSV, TXT, and MD. Multiple extraction backends — Fast (text-only), Docling (multimodal), and NV-Ingest (NVIDIA NIM). Outputs RAG-ready vector embeddings.
Data Studio
02Clean, deduplicate, augment, and format your training data. Built-in quality checks, bias detection, and dataset versioning ensure your models are trained on reliable, representative data.
Model Hub
03Choose from a curated catalog of base models — Llama, Mistral, Qwen, DeepSeek, Nemotron — or bring your own. Automated benchmarking helps you select the right architecture for your domain.
Training
04Fine-tune using NVIDIA NeMo, Unsloth, or MLX — whichever framework is optimal for your use case. PEFT, LoRA, QLoRA, full fine-tuning — Forge selects the right approach for your data and compute budget.
Evaluation
05Seven benchmark dimensions: accuracy, latency, safety, bias, regulatory alignment, robustness, and domain coverage. Automated evaluation suites produce quantitative scores and detailed failure analysis.
Packaging
06Production-ready deployment packages: NIM containers, GGUF quantized models, or Python wheels. Each package includes model cards, deployment guides, and cryptographic integrity receipts.
Multi-backend extraction
Choose the extraction backend that fits your format, quality, and licensing requirements. From fast text-only to NVIDIA NIM-powered multimodal extraction.
| Backend | License | Formats | Status |
|---|---|---|---|
| Fast (text-only) | Apache 2.0 | PDF, DOCX, TXT, MD, HTML | ready |
| Docling (multimodal) | Apache 2.0 | PDF, DOCX, PPTX, Images | available |
| NV-Ingest (NVIDIA NIM) | NVIDIA AI Enterprise | PDF, DOCX, Images, Tables | license required |
Built on the full NVIDIA AI platform
Every component of Forge runs on NVIDIA infrastructure — from training frameworks to inference engines to deployment containers.
NeMo
Model customization framework
NIM
Optimized inference microservices
TensorRT-LLM
LLM inference acceleration
Megatron-Core
Large-scale training
Milvus
Vector database
NV-EmbedQA
Embedding models
Three deployment targets
Ship your vertical AI product the way your customers need it — as optimized containers, local models, or embedded libraries.
NIM Container
NVIDIA-optimized inference container with TensorRT-LLM acceleration. Enterprise-ready with built-in security, monitoring, and API compatibility. Deploy on DGX, cloud GPU instances, or your own hardware.
GGUF (Quantized)
Local-first deployment with quantized model formats. Run on consumer hardware, edge devices, or air-gapped environments. Multiple quantization levels for precision-vs-performance tradeoffs.
Python Wheel
Lightweight distribution for embedding into existing Python applications. Import your model as a library — no container orchestration required. Ideal for CI/CD and automated workflows.
Every artifact, verifiably yours
Forge doesn't just build models — it builds a chain of cryptographic evidence that proves what was built, how it was trained, and how it performed. Regulators and auditors get the receipts they need.
Ed25519 Receipts
Every model version, every training run, every evaluation result is signed with Ed25519. Cryptographic proof that your model is exactly what was trained and tested.
Merkle-Batched Audit Trails
Immutable, tamper-evident logs of the entire build pipeline. Merkle tree batching provides efficient verification while maintaining cryptographic integrity across millions of pipeline events.
Bring your expertise. We'll build the product.
Forge is in early access. If you have deep domain expertise and want to build a vertical AI product, let's talk.
Request early access