✳ THE FORGEDATA PLATFORM

Everything you need to make a model your own.

A single platform for dataset preparation, GPU accelerated fine-tuning, evaluation, and inference. Keep the context behind every model version, from its source data to its API endpoint.

YOUR NEXT MODEL STARTS HEREInteractive product preview
Workspace / Training jobs Preview environment
PROJECT / CUSTOMER SUPPORT

Monitor your training run.

Sample project
support-assistant-v1Meta Llama 3.1 8B · QLoRA
Training
Training loss0.284 ↓ 82.4%
Epoch2 / 3
GPUNVIDIA A100
Elapsed time00:24:18
Training loss train_loss eval_loss
0200400600800 steps
Training in progress68% complete
LoRA & QLoRA training Your data stays yours API deployment

Product preview. Capabilities described below are the target early access scope; availability will be confirmed during onboarding.

Your datasets, ready for training

Import CSV and JSONL files into a shared workspace. Review schemas, identify invalid examples, and reserve a holdout set before training.

CSV & JSONLDataset versioningValidation

Open models. Your starting point.

Choose from a planned catalog of Llama, Mistral, Qwen, and Gemma models. Select a foundation that fits your task, model license, and compute budget.

Model catalogLicense visibilityAdapter export

Efficient adaptation with LoRA

Configure low rank adapters and quantized training with QLoRA. Set hyperparameters, save configurations, and compare experiments in one place.

LoRA / QLoRAHyperparametersCheckpoints

See every step of training

Track loss curves, epochs, GPU utilization, and job status. Review evaluation results alongside the base model before choosing a checkpoint.

Job monitoringEvaluationExperiment comparison

A shorter route to deployment

The deployment workflow turns an evaluated model into an inference endpoint. Keep model versions connected to the data and jobs that created them.

Model versioningEndpoint managementUsage visibility

An API your application understands

Integrate with a familiar chat completions pattern. Keep model calls separate from application logic so your team can iterate on either independently.

API accessStreaming roadmapSDK compatibility goal
✳ OPEN BY DESIGN

Choose your foundation.

Start with an open model that fits your task. The initial catalog is planned around these model families; exact versions and configurations will be confirmed during early access.

Model familyAdaptation workflowBring your own dataAvailability
Meta LlamaLoRA / QLoRACSV and JSONLPlanned catalog
MistralLoRA / QLoRACSV and JSONLPlanned catalog
QwenLoRA / QLoRACSV and JSONLPlanned catalog
GemmaLoRA / QLoRACSV and JSONLPlanned catalog

Model availability depends on architecture compatibility and license terms. Adapter and model usage remain subject to each foundation model’s license.

✳ ACCELERATED BY DESIGN

Built around NVIDIA GPU workloads.

Training and inference are at the heart of the product architecture. Here is how we plan to use the accelerated computing stack.

CUDA

Architecture target

GPU accelerated tensor operations for model training and serving.

TensorRT-LLM

Planned integration

Optimized inference execution for supported model architectures.

NVIDIA NIM

Planned integration

Containerized model serving as a future deployment option.

NVIDIA NeMo

Under evaluation

Training and evaluation tooling for reproducible model development.

Technology roadmap, not a claim of certification, partnership, or current integration. No NVIDIA affiliation is implied.

YOUR NEXT CHAPTER IN AI

From your first dataset.
To your next release.

Bring a use case. Explore a model workflow built for your team.

Request access