Topic

LLM Fine-tuningの論文・実装まとめ

LoRA、RLHF、SFT、軽量ファインチューニングなど、LLMを用途に合わせる技術を扱います。

まず見るべき実装候補

461 articles

新着記事

githubGitHubあり2026-11-04

wandb — The AI developer platform. Use Weights & Biases to train and fine-tune models, and manage models from experimentation to production.

Weights & Biasesは、AI開発を支援するプラットフォームです。このプラットフォームは、モデル開発から生産準備までを支援し、コストをコントロールし、モデルとデータへのアクセスを管理します。

自然言語処理ファインチューニング
用途
AI開発プラットフォーム
難易度
Easy
コスト
Medium
→
githubGitHubあり2026-11-04

paperless-ngx — A community-supported supercharged document management system: scan, index and archive all your documents

paperless-ngxは、コミュニティによってサポートされたスーパーチャージドのドキュメント管理システムで、ドキュメントのスキャン・インデックス・アーカイブが可能である。

自然言語処理大規模言語モデル分類テキスト
用途
ドキュメント管理
難易度
Easy
コスト
High
→
githubGitHubあり2026-11-04

unsloth — Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.

Unsloth Studioは、オープンモデルのトレーニングと実行を支援するWebUIです。このライブラリは、Gemma4、Qwen3.5などのオープンモデルのテストとトレーニングを支援するために使われます。

自然言語処理大規模言語モデル生成画像テキスト
用途
オープンモデルのトレーニングと実行
難易度
Easy
コスト
High
→
githubGitHubあり2026-11-04

ART — Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.6, GPT-OSS, Llama, and more!

ARTは、多段強化学習トレーナーです。このトレーナーは、GRPOを使用して、現実世界のタスクに対して、多段強化学習を行うことができます。

自然言語処理大規模言語モデル強化学習
用途
多段強化学習トレーナー
難易度
Easy
コスト
High
→
githubGitHubあり2026-11-04

mlflow — The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while controlling costs and managing access to models and data.

このリポジトリでは、AIワークロードを管理するためのシステムであるSkypilotを提供しています。

品質予測/異常検知自然言語処理大規模言語モデル
用途
AIワークロードを管理するためのシステム
難易度
Easy
コスト
High
→
githubGitHubあり2026-11-04

skypilot — The AI Compute Platform for frontier teams. SkyPilot turns fragmented AI compute into one AI supercomputer, so frontier AI teams build custom intelligence faster.

このリポジトリでは、AIアプリケーションをローカルに実行できるツールキットであるRunAnywhere-sdksを提供しています。

自然言語処理大規模言語モデル
用途
AIアプリケーションをローカルに実行できるツールキット
難易度
Easy
コスト
High
→
githubGitHubあり2026-11-04

lance — Open Lakehouse Format for Multimodal AI. Convert from Parquet in 2 lines of code for 100x faster random access, vector index, and data versioning. Compatible with Pandas, DuckDB, Polars, Pyarrow, and PyTorch with more integrations coming..

マルチモーダルAIに適したオープンレイクハウスフォーマットです。このフォーマットでは、パレットからデータを2行のコードで変換することができ、100倍速くなります。また、ベクトルインデックスやデータバージョニングが可能です

自然言語処理大規模言語モデルマルチモーダル
用途
オープンレイクハウスフォーマット
難易度
Easy
コスト
High
→