> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/automodel/latest/model-coverage/large-language-models/moonshotai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/automodel/_mcp/server.

# Moonshot AI

> Browse NeMo AutoModel coverage for Moonshot AI large language models.

* [`Kimi-K2-Base`](/model-coverage/large-language-models/moonshotai/Kimi-K2-Base)
* [`Kimi-K3`](/model-coverage/large-language-models/moonshotai/Kimi-K3)
* [`Kimi-Linear-48B-A3B-Instruct`](/model-coverage/large-language-models/moonshotai/Kimi-Linear-48B-A3B-Instruct)
* [`Moonlight-16B-A3B`](/model-coverage/large-language-models/moonshotai/Moonlight-16B-A3B)

> Browse NeMo AutoModel coverage for Moonshot AI large language models.

## Docs

- [Kimi-K2-Base](https://docs.nvidia.com/nemo/automodel/latest/model-coverage/large-language-models/moonshotai/Kimi-K2-Base.md): Use NeMo AutoModel to fine-tune Kimi-K2-Base for LLM workflows with checked-in text generation recipes and documented distributed settings.
- [Kimi-K3](https://docs.nvidia.com/nemo/automodel/latest/model-coverage/large-language-models/moonshotai/Kimi-K3.md): Fine-tune Kimi-K3 for text generation with NeMo AutoModel using checked-in multi-node recipes with expert, pipeline, and context parallelism.
- [Kimi-Linear-48B-A3B-Instruct](https://docs.nvidia.com/nemo/automodel/latest/model-coverage/large-language-models/moonshotai/Kimi-Linear-48B-A3B-Instruct.md): Use Kimi-Linear-48B-A3B-Instruct with NeMo AutoModel for language model fine-tuning, with documented checkpoints, runnable recipes, setup, and model reference details.
- [Moonlight-16B-A3B](https://docs.nvidia.com/nemo/automodel/latest/model-coverage/large-language-models/moonshotai/Moonlight-16B-A3B.md): Fine-tune Moonlight-16B-A3B with NeMo AutoModel on 8 GPUs using FSDP2, 8-way expert parallelism, Transformer Engine, and DeepEP.