Ministral-3-8B-Reasoning-2512 | NVIDIA NeMo AutoModel

Ministral-3-8B-Reasoning-2512 has a checked-in NeMo AutoModel recipe for image-text-to-text. The Hugging Face configuration declares the Mistral3ForConditionalGeneration architecture.

Fine-Tune Ministral-3-8B-Reasoning-2512

Follow the installation instructions, then run the recipe from the repository root:

automodel examples/vlm_finetune/mistral/ministral3_8b_medpix.yaml --nproc-per-node 8

Choose a Workflow

Goal	Start Here
Run the primary recipe for this model	Use the recipe configuration.
Prepare your environment	Follow the installation instructions.

Configuration

Setting	Configuration
Hardware	-
Strategy	FSDP2; tp_size=1, cp_size=1
Nodes	-
Features	Checkpointing (enabled=true)
Advanced	torch.bfloat16

Model Reference

Model Architecture

Property	Value
Task	Image-text-to-text
Architecture	`Mistral3ForConditionalGeneration`

Available Models

mistralai/Ministral-3-8B-Reasoning-2512

Vision-language model fine-tuning guide