{"meta":{"title":"AI モデルの比較","intro":"Copilot Chatで使用可能な AI モデルを比較し、タスクに最適なモデルを選択します。","product":"GitHub Copilot","breadcrumbs":[{"href":"/ja/copilot","title":"GitHub Copilot"},{"href":"/ja/copilot/reference","title":"リファレンス"},{"href":"/ja/copilot/reference/ai-models","title":"AI モデル"},{"href":"/ja/copilot/reference/ai-models/model-comparison","title":"モデルの比較"}],"documentType":"article"},"body":"# AI モデルの比較\n\nCopilot Chatで使用可能な AI モデルを比較し、タスクに最適なモデルを選択します。\n\n## の AI モデルの比較 GitHub Copilot\n\nGitHub Copilot では、さまざまな機能を持つ複数の AI モデルがサポートされています。 選択したモデルは、インライン提案の Copilot Chat と Copilot による応答の品質と関連性に影響します。 待機時間が少ないモデルや、特定のタスクではハルシネーションが少ない、またはパフォーマンスが高いモデルがあります。 このガイドは、モデル名だけでなく、タスクに基づいて最適なモデルを選ぶのに役立ちます。\n\n> \\[!NOTE]\n>\n> * 異なるモデルでは、トークンの価格に基づいて異なるレートで AI credits が使用されます。 詳細については、「[GitHub Copilot のモデルと価格設定](/ja/copilot/reference/copilot-billing/models-and-pricing)」を参照してください。\n> * モデルを選択するときに **\\[自動]** オプションを選択すると、モデルの可用性とタスクの複雑さに基づいて、タスクに最適なモデルが自動的に選択されます。  有料 Copilot プランをご利用の場合は、\\[ **自動]** オプションを使用すると、モデル コストの割引を受けることができます。 「[について Copilotauto model selection](/ja/copilot/concepts/models/auto-model-selection)」を参照してください。\n\n### タスク別の推奨モデル\n\n次の表は、適切なモデルをすばやく見つけるのにお使いください。詳細については、後のセクションを参照してください。\n\n| モデル        | タスク領域 | 長所 (主なユース ケース) | 参考資料 |\n| ---------- | ----- | -------------- | ---- |\n|            |       |                |      |\n| GPT-5 mini |       |                |      |\n\n```\n                               | General-purpose coding and writing | Fast, accurate code completions and explanations        | [GPT-5 mini model card](https://cdn.openai.com/gpt-5-system-card.pdf) |\n```\n\n\\|  |\n\\| GPT-5.3-Codex\n\\| Agentic software development | Agentic tasks        | [GPT-5.3-Codex model card](https://deploymentsafety.openai.com/gpt-5-3-codex) |\n\\|  |\n\\| GPT-5.4\n\\| Deep reasoning and debugging | Multi-step problem solving and architecture-level code analysis        | [GPT-5.4 model card](https://deploymentsafety.openai.com/gpt-5-4-thinking/introduction) |\n\\|  |\n\\| GPT-5.4 mini\n\\| Agentic software development | Codebase exploration and is especially effective when using grep-style tools        | [GPT-5.4 mini model card](https://deploymentsafety.openai.com/gpt-5-4-thinking/gpt-5-4-thinking.pdf) |\n\\|  |\n\\| GPT-5.5\n\\| Deep reasoning and debugging | Multi-step problem solving and architecture-level code analysis        | [GPT-5.5 model card](https://deploymentsafety.openai.com/gpt-5-5/gpt-5-5.pdf) |\n\\|  |\n\\| GPT-5.6 Luna\n\\| Fast help with simple or repetitive tasks | Quick, cost-efficient responses for smaller, faster coding tasks        | [GPT-5.6 model card](https://deploymentsafety.openai.com/gpt-5-6/gpt-5-6.pdf) |\n\\|  |\n\\| GPT-5.6 Sol\n\\| Deep reasoning and debugging | Complex reasoning over large codebases and long-running agentic work        | [GPT-5.6 model card](https://deploymentsafety.openai.com/gpt-5-6/gpt-5-6.pdf) |\n\\|  |\n\\| GPT-5.6 Terra\n\\| General-purpose coding and agent tasks | Balanced everyday interactive and agentic coding        | [GPT-5.6 model card](https://deploymentsafety.openai.com/gpt-5-6/gpt-5-6.pdf) |\n\\|  |\n\\| GPT-6 Astra\n\\| Long-horizon, autonomous coding and agentic tasks | Long-horizon coding tasks with continuous planning, batched diagnosis and verification, and independent result confirmation        | [GPT-6 Astra model card](https://deploymentsafety.openai.com/gpt-6-astra/gpt-6-astra.pdf) |\n\\|  |\n\\| Claude Fable 5\n\\| Long-horizon, autonomous coding and knowledge-work | First-attempt correctness through upfront reasoning, aggressive parallel tool batching, and proactive verification of pre-existing test state before debugging        | [Claude Fable 5 model card](https://anthropic.com/claude-fable-5-mythos-5-system-card) |\n\\|  |\n\\| Claude Fable 5.1\n\\| Long-horizon, autonomous coding and knowledge-work | Substantial, long-running coding tasks, including deep codebase research, feature development, and complex agentic workflows        | [Claude Fable 5.1 model card](https://www-cdn.anthropic.com/0339e6a7c5c7b87f5c07798616dc32c215d14235/Claude%20Fable%205.1%20&%20Claude%20Mythos%205.1%20System%20Card.pdf) |\n\\|  |\n\\| Claude Haiku 4.5\n\\| Fast help with simple or repetitive tasks | Fast, reliable answers to lightweight coding questions        | [Claude Haiku 4.5 model card](https://assets.anthropic.com/m/99128ddd009bdcb/Claude-Haiku-4-5-System-Card.pdf) |\n\\|  |\n\\| Claude Opus 4.7\n\\| Deep reasoning and debugging | Complex problem-solving challenges, sophisticated reasoning        | [Claude Opus 4.7 model card](https://cdn.sanity.io/files/4zrzovbb/website/037f06850df7fbe871e206dad004c3db5fd50340.pdf) |\n\\|  |\n\\| Claude Opus 4.8\n\\| Deep reasoning and debugging | Complex problem-solving challenges, sophisticated reasoning        | [Claude Opus 4.8 model card](https://www-cdn.anthropic.com/0b4915911bb0d19eca5b5ee635c80fef830a37ea.pdf) |\n\\|  |\n\\| Claude Opus 4.8 (fast mode) (preview)\n\\| Deep reasoning and debugging | Complex problem-solving challenges, sophisticated reasoning        | [Claude Opus 4.8 model card](https://www-cdn.anthropic.com/0b4915911bb0d19eca5b5ee635c80fef830a37ea.pdf) |\n\\|  |\n\\| Claude Opus 5\n\\| Deep reasoning and debugging | Complex problem-solving challenges, sophisticated reasoning        | [Claude Opus 5 model card](https://www-cdn.anthropic.com/c5fbac3f0b1280a933ebd26d3cb8bb9f5bdeaf48/Claude%20Opus%205%20System%20Card.pdf) |\n\\|  |\n\\| Claude Sonnet 4.6\n\\| General-purpose coding and agent tasks | Complex problem-solving challenges, sophisticated reasoning        | [Claude Sonnet 4.6 model card](https://www-cdn.anthropic.com/78073f739564e986ff3e28522761a7a0b4484f84.pdf) |\n\\|  |\n\\| Claude Sonnet 5\n\\| General-purpose coding and agent tasks | Complex problem-solving challenges, sophisticated reasoning        | [Claude Sonnet 5 model card](https://www-cdn.anthropic.com/9e6a1044980d8c4ed85669faf9c2a8342e2e9f1e/Claude%20Sonnet%205%20System%20Card.pdf) |\n\\|  |\n\\| Gemini 3.5 Flash\n\\| Fast help with simple or repetitive tasks | Fast, reliable answers to lightweight coding questions        | [Gemini 3.5 Flash model card](https://storage.googleapis.com/deepmind-media/Model-Cards/Gemini-3-5-Flash-Model-Card.pdf) |\n\\|  |\n\\| Gemini 3.6 Flash\n\\| Fast help with simple or repetitive tasks | Fast, reliable answers to lightweight coding questions        | [Gemini 3.6 Flash model card](https://storage.googleapis.com/deepmind-media/Model-Cards/Gemini-3-6-Flash-Model-Card.pdf) |\n\\|  |\n\\| Gemini 3.7 Flash\n\\| Fast help with simple or repetitive tasks | Fast, reliable answers to lightweight coding questions        | [Gemini 3.7 Flash model card](https://storage.googleapis.com/deepmind-media/Model-Cards/Gemini-3-7-Flash-Model-Card.pdf) |\n\\|  |\n\\| Gemini 3.8 Flash\n\\| Fast help with simple or repetitive tasks | Fast, reliable answers to lightweight coding questions        | [Gemini 3.8 Flash model card](https://storage.googleapis.com/deepmind-media/Model-Cards/Gemini-3-8-Flash-Model-Card.pdf) |\n\\|  |\n\\| MAI-Code-1-Flash\n\n[^mai-code-1-flash]                                   | General-purpose coding and writing | Fast, accurate code completions and explanations        | [MAI-Code-1-Flash model card](https://aka.ms/mai-code-1-flash-modelcard) |\n\\|  |\n\\| MAI-Code-1.1-Flash\n\n[^mai-code-1-flash]                                   | General-purpose coding and writing, image understanding | Fast code completions and explanations, instruction following, tool use        | [MAI-Code-1.1-Flash model card](https://microsoft.ai/pdf/MAI-Code-1.1-Flash-Model-Card.PDF) |\n\\|  |\n\\| Grok 4.5\n\\| General-purpose coding and agent tasks | Complex problem-solving challenges, sophisticated reasoning        | [Grok 4.5 model card](https://media.x.ai/v1/website/card-7f81d41b.pdf) |\n\\|  |\n\\| Grok 4.6\n\\| General-purpose coding and agent tasks | Complex problem-solving challenges, sophisticated reasoning        | [Grok 4.6 model card](https://media.x.ai/v1/website/card-4p6-4cd2dc57.pdf) |\n\\|  |\n\\| Qwen2.5\n\\| General-purpose coding and writing | Code generation, reasoning, and code repair / debugging        | [Qwen2.5 model card](https://arxiv.org/pdf/2409.12186) |\n\\|  |\n\\| Kimi K2.7 Code\n\\| General-purpose coding and agent tasks | Fast, reliable answers to lightweight coding questions        | [Kimi K2.7 Code model card](https://huggingface.co/moonshotai/Kimi-K2.7-Code) |\n\\|  |\n\\| Kimi K3\n[^kimi-k3]                                   | Agentic coding and long-context work | Multi-step agent tasks across large codebases        | [Kimi K3 model card](https://huggingface.co/moonshotai/Kimi-K3) |\n\\|                                        |\n\n## タスク: 汎用のコーディングとライティング\n\n品質、速度、コスト効率のバランスが必要な一般的な開発タスクには、これらのモデルを使います。 これらのモデルは、特定の要件がない場合に既定で使うのに適しています。\n\n| モデル              | 適している理由                                                                  |\n| ---------------- | ------------------------------------------------------------------------ |\n| GPT-5.3-Codex    | 機能、テスト、デバッグ、リファクタリング、レビューなどの複雑なエンジニアリング タスクに関する高品質のコードを、長い指示なしで提供します。    |\n| GPT-5 mini       | コーディングとライティングのほとんどのタスクに対する信頼性の高い既定値。 高速かつ正確で、言語やフレームワークが違っても良好に動作します。    |\n| GPT-5.6 Terra    | 日常のインタラクティブでエージェント的なコーディングのためのバランスの取れたオールラウンドの選択。                        |\n| Raptor mini      | 迅速で正確なインライン提案と説明に特化しています。                                                |\n| MAI-Code-1-Flash | 強力な命令フォローとアダプティブ推論により、日常的なコーディング タスク、書き込み、マルチターン開発ワークフローの信頼性の高い既定値になります。 |\n\n### これらのモデルを使用すべき場合\n\n次のことが必要な場合は、これらのモデルのいずれかをお使いください。\n\n* 関数、短いファイル、またはコードの差分を記述または確認します。\n* ドキュメント、コメント、または概要を生成します。\n* エラーや予期しない動作をすばやく説明します。\n* 英語以外のプログラミング環境で作業します。\n\n### 別のモデルを使用すべきとき\n\n複雑なリファクタリング、アーキテクチャの決定、またはマルチステップ ロジックの作業を行っている場合は、「[詳細な推論とデバッグ](#task-deep-reasoning-and-debugging)」のモデルを検討してください。 反復的な編集や 1 回限りのコードの提案など、高速でシンプルなタスクの場合は、「[単純なタスクや反復的なタスクに関する高速の支援](#task-fast-help-with-simple-or-repetitive-tasks)」を参照してください。\n\n## タスク: 単純なタスクや反復的なタスクに関する高速の支援\n\nこれらのモデルは、速度と応答性のために最適化されています。 これらは簡単な編集、ユーティリティ関数、構文のヘルプ、軽量のプロトタイプ作成に最適です。 必要以上に詳細な、または長い推論チェーンを待たずに、速やかに回答を得ることができます。\n\n### 推奨モデル\n\n| モデル              | 適している理由                                                           |\n| ---------------- | ----------------------------------------------------------------- |\n| GPT-5.6 Luna     | より小さく、より速いタスクのための軽量でコスト効率の高いオプション。 GPT-5.6 ファミリの最も低コストのモデル。       |\n| Claude Haiku 4.5 | 迅速な応答と高品質の出力のバランスを取ります。 小さなタスクや軽量のコードの説明に最適です。                    |\n| MAI-Code-1-Flash | アダプティブ効率で迅速なコーディング タスクを処理し、単純な要求に対して簡潔に保ち、不要な深さなしで高速で正確な応答を提供します。 |\n\n### これらのモデルを使用すべき場合\n\n次のことが必要な場合は、これらのモデルのいずれかをお使いください。\n\n* 小さい関数やユーティリティ コードを記述または編集します。\n* 構文や言語について簡単な質問をします。\n* 最小限のセットアップでアイデアのプロトタイプを作成します。\n* 簡単なプロンプトや編集に関するフィードバックをすばやく入手します。\n\n### 別のモデルを使用すべきとき\n\n複雑なリファクタリング、アーキテクチャの決定、またはマルチステップ ロジックの作業を行っている場合は、「[詳細な推論とデバッグ](#task-deep-reasoning-and-debugging)」のモデルを検討してください。\nより強力な汎用の推論またはより構造化された出力を必要とするタスクについては、「[汎用コーディングとライティング](#task-general-purpose-coding-and-writing)」を参照してください。\n\n## タスク: 詳細な推論とデバッグ\n\nこれらのモデルは、ステップ バイ ステップの推論、複雑な意思決定、または高コンテキストの認識を必要とするタスク用に設計されています。 構造化された分析、よく考えられたコード生成、または複数のファイルの理解が必要な場合に適切に機能します。\n\n### 推奨モデル\n\n| モデル                        | 適している理由                                                                    |\n| -------------------------- | -------------------------------------------------------------------------- |\n| GPT-5 mini                 | GPT-5 よりも高速な応答と低いリソース使用量で、深い推論とデバッグを実現します。 対話型セッションやステップバイステップのコード分析に最適です。 |\n| GPT-5.5                    | 複雑な推論、コード分析、技術的な意思決定に優れています。                                               |\n| GPT-5.6 Sol                | GPT-5.6 ファミリーで最も高い推論の上限。 大規模なコードベースに対する複雑な推論や、要求の厳しい実行時間の長いエージェント作業に最適です。  |\n| Claude Sonnet 4.6          | より信頼性の高い補完と負荷の下でのよりスマートな推論でSonnet 4.5を改善します。                               |\n| Claude Opus 4.7            | Anthropic の最も強力なモデルです。                                                     |\n| Claude Opus 4.6によって改善されます。 |                                                                            |\n| Gemini 3.1 Pro             | 長いコンテキストでの高度な推論と、科学的または技術的な分析。                                             |\n\n### これらのモデルを使用すべき場合\n\n次のことが必要な場合は、これらのモデルのいずれかをお使いください。\n\n* 複数のファイルを対象にコンテキストに関する複雑な問題をデバッグします。\n* 大規模な、または相互に接続されたコードベースをリファクタリングします。\n* 複数のレイヤーを対象とする機能またはアーキテクチャを計画します。\n* ライブラリ、パターン、またはワークフローの間のトレードオフを比較します。\n* ログ、パフォーマンス データ、またはシステムの動作を分析します。\n\n### 別のモデルを使用すべきとき\n\n高速の反復的または軽量なタスクについては、「[単純なタスクや反復的なタスクに関する高速の支援](#task-fast-help-with-simple-or-repetitive-tasks)」を参照してください。\n一般的な開発ワークフローまたはコンテンツの生成については、「[汎用のコーディングとライティング](#task-general-purpose-coding-and-writing)」を参照してください。\n\n## タスク: ビジュアルの処理 (図、スクリーンショット)\n\nスクリーンショット、図、UI コンポーネント、その他の視覚的な入力について質問したい場合は、これらのモデルを使います。 これらのモデルはマルチモーダル入力をサポートしており、フロントエンドの作業や視覚的なデバッグに適しています。\n\n| モデル               | 適している理由                                                                                             |\n| ----------------- | --------------------------------------------------------------------------------------------------- |\n| GPT-5 mini        | コーディングとライティングのほとんどのタスクに対する信頼性の高い既定値。 高速かつ正確で、視覚的な推論タスクにおいてマルチモーダル入力に対応します。 言語やフレームワークが違っても良好に動作します。 |\n| Claude Sonnet 4.6 | より信頼性の高い補完と負荷の下でのよりスマートな推論でSonnet 4.5を改善します。                                                        |\n| Gemini 3.1 Pro    | 詳細な推論とデバッグは、複雑なコード生成、デバッグ、調査ワークフローに最適です。                                                            |\n\n### これらのモデルを使用すべき場合\n\n次のことが必要な場合は、これらのモデルのいずれかをお使いください。\n\n* 図、スクリーンショット、または UI コンポーネントについて質問します。\n* 視覚的な下書きまたはワークフローに関するフィードバックを受け取ります。\n* ビジュアル コンテキストからフロントエンドの動作を理解します。\n\n> \\[!TIP]\n> 画像入力をサポートしていないコンテキストでモデルを使っている場合 (コード エディターなど)、視覚的な推論の利点は見られません。 MCP サーバーを使って、視覚的な入力に間接的にアクセスできる場合があります。 「[モデル コンテキスト プロトコル (MCP) サーバーを使用した GitHub Copilot Chatの拡張](/ja/copilot/how-tos/provide-context/use-mcp-in-your-ide/extend-copilot-chat-with-mcp)」を参照してください。\n\n### 別のモデルを使用すべきとき\n\nタスクに詳細な推論または大規模なリファクタリングが含まれている場合は、「[詳細な推論とデバッグ](#task-deep-reasoning-and-debugging)」のモデルを検討してください。 テキストのみのタスクや簡単なコード編集の場合は、「[単純なタスクや反復的なタスクに関する高速の支援](#task-fast-help-with-simple-or-repetitive-tasks)」を参照してください。\n\n## モデル固有の考慮事項\n\n一部のモデルには、タスクに対して選択する前に理解するのに役立つ動作、制限、またはセーフガードがあります。\n\n### Kimi K3\n\nKimi K3 は、長いコンテキスト、マルチステップのコーディング、およびエージェントワークフロー用に設計されています。 プレリリース テストでは、このモデルは特定のリスクの高いプロンプトに対して高いリスクを示し、機密性の高いトピックに関連する要求を拒否する他のモデルよりも一貫性が低かった。 これらの動作は、トレーニング後のモデルの安全性と配置の違いを反映している可能性があります。 特定されたリスクを軽減するために、 GitHub Copilot に追加のセーフガードをデプロイしました。 他のモデルと同様に、企業は、特定のユース ケースと要件に照らして、モデルの機能、制限、保護を評価する必要があります。\n\n## 次のステップ\n\n[^mai-code-1-flash]: MAI models are continuously improving models. Performance and behavior may evolve over time as new checkpoints are released.\n\n[^kimi-k3]: For important information about Kimi K3 behavior and safeguards, see [Model-specific considerations](#kimi-k3).\n\n適切なモデルを選択すると、 Copilotを最大限に活用できます。 使用するモデルがわからない場合は、 GPT-5 miniなどの汎用オプションから始めて、ニーズに応じて調整します。\n\n* モデルの仕様と価格について詳しくは、「[GitHub Copilotでサポートされている AI モデル](/ja/copilot/reference/ai-models/supported-models)」をご覧ください。\n* さまざまなモデルの使用方法の他の例については、「[さまざまなタスクを使った AI モデルの比較](/ja/copilot/tutorials/compare-ai-models)」をご覧ください。\n* モデルの切り替えについては、「[GitHub Copilot Chat の AI モデルの変更](/ja/copilot/how-tos/use-ai-models/change-the-chat-model)」または「[GitHub Copilot インライン提案の AI モデルを変更する](/ja/copilot/how-tos/use-ai-models/change-the-completion-model)」をご覧ください。\n* Copilot Chatがさまざまな AI モデルを提供する方法については、[GitHub Copilotのモデルのホスティング](/ja/copilot/reference/ai-models/model-hosting) を参照してください。"}