Kimi-K3

View as Markdown

Kimi-K3 is a Moonshot AI vision-language model with a MoonViT3d vision tower and a Kimi Linear Mixture-of-Experts language backbone.

Set up NeMo AutoModel with the latest container or follow the installation instructions.

Fine-Tune Kimi-K3

From the repository root, run:

uv run automodel --nproc-per-node=8 examples/llm_finetune/kimi/k3_hellaswag.yaml

Choose a Workflow

GoalStart Here
Supervised fine-tuning (SFT) - text-only Kimi-K3 with EP32 + PP8Use k3_hellaswag.yaml. Dataset: HellaSwag.

Model Reference

Model Architecture

PropertyValue
TaskImage-Text-to-Text / Text Generation
ArchitecturesKimiK3ForConditionalGeneration, KimiK3ForCausalLM
Hugging Face Organizationmoonshotai
  • KimiK3ForConditionalGeneration combines the MoonViT3d vision tower and multimodal projector with the Kimi Linear language backbone.
  • KimiK3ForCausalLM exposes the text-only Kimi Linear backbone.

Available Models

ModelHF ID
Kimi-K3moonshotai/Kimi-K3