Loading...Working on your request
モデル候補リスト

Best long-context model APIs for large documents

Compare long-context model APIs by context window, price, model source, and recommended document-heavy use cases.

モデルを見るコストを見積もる

この候補リストは何に使う？

Long-context models are useful when prompts include full contracts, knowledge-base exports, support histories, or large code files. The tradeoff is that longer prompts can quickly increase cost, so teams should compare both context window and input price before shipping.

ソース基準: NextModel curated catalog and OpenRouter context metadata when available.

Context

推奨候補 long-context models

まず候補リストから始め、実際のプロンプトで試し、本番ルーティング前に月額コストを比較します。

GoogleCatalog

Google: Gemini 2.5 Pro

91

Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy...

$1.25 / 1M tokensInput$10 / 1M tokensOutput1MContext
Best forlong-context analysis, vision workflows, scientific reasoning
RoutingConfigured
Tool callingVisionJSON modeLong contextReasoningStreaming
OpenRouter if availableOpenRouter public Models API live metadata; public price comes from the registry pricing rule
View details
DeepSeekCatalog

DeepSeek: DeepSeek V4 Flash

88

DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...

$0.112 / 1M tokensInput$0.224 / 1M tokensOutput1MContext
Best forlow-cost Chinese tasks, long-context summary, batch code assistance
RoutingConfigured
Tool callingJSON modeLong contextReasoningLow cost
OpenRouter if availableOpenRouter public Models API live metadata; public price comes from the registry pricing rule
View details
GoogleCatalog

Google: Gemini 2.5 Flash

86

Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in "thinking" capabilities, enabling it to provide responses with greater...

$0.3 / 1M tokensInput$2.50 / 1M tokensOutput1MContext
Best forlong-document summarization, image Q&A, fast multimodal routing
RoutingConfigured
Tool callingVisionJSON modeLong contextStreamingLow cost
OpenRouter if availableOpenRouter public Models API live metadata; public price comes from the registry pricing rule
View details
MetaCatalog

Meta: Llama 4 Maverick

81

Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward...

$0.15 / 1M tokensInput$0.6 / 1M tokensOutput1MContext
Best foropen-model workflows, cost-sensitive long context, classification
RoutingConfigured
JSON modeLong contextStreamingLow costTool callingVision
OpenRouter if availableOpenRouter public Models API live metadata; public price comes from the registry pricing rule
View details

比較表

価格、プロバイダー、コンテキスト、機能、ソースで候補を比較します。

本番候補を絞り込むとき、フォールバック方針を作るとき、モデル経済性を比べるときに使います。

ModelProviderInputOutputContextCapabilitiesBest forLatencyStatusSource
Google: Gemini 2.5 Progoogle/gemini-2.5-proGoogle$1.25 / 1M tokens$10 / 1M tokens1M
Tool callingVisionJSON modeLong context
long-context analysis, vision workflows1500-5000msCatalogOpenRouter if available
DeepSeek: DeepSeek V4 Flashdeepseek/deepseek-v4-flashDeepSeek$0.112 / 1M tokens$0.224 / 1M tokens1M
Tool callingJSON modeLong contextReasoning
low-cost Chinese tasks, long-context summary800-2600msCatalogOpenRouter if available
Google: Gemini 2.5 Flashgoogle/gemini-2.5-flashGoogle$0.3 / 1M tokens$2.50 / 1M tokens1M
Tool callingVisionJSON modeLong context
long-document summarization, image Q&A900-2800msCatalogOpenRouter if available
Meta: Llama 4 Maverickmeta-llama/llama-4-maverickMeta$0.15 / 1M tokens$0.6 / 1M tokens1M
JSON modeLong contextStreamingLow cost
open-model workflows, cost-sensitive long context950-2800msCatalogOpenRouter if available
Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7Anthropic$5 / 1M tokens$25 / 1M tokens1M
Tool callingJSON modeLong contextReasoning
frontier reasoning, large codebase review2300-6800msCatalogOpenRouter if available
Anthropic: Claude Sonnet 4.5anthropic/claude-sonnet-4.5Anthropic$3 / 1M tokens$15 / 1M tokens1M
Tool callingJSON modeLong contextReasoning
coding agents, code review1600-4800msCatalogOpenRouter if available
Qwen: Qwen3 Coder Plusqwen/qwen3-coder-plusAlibaba Cloud / Qwen$0.65 / 1M tokens$3.25 / 1M tokens1M
Tool callingJSON modeLong contextStreaming
Chinese engineering workflows, code generation1200-3900msCatalogOpenRouter if available
MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6Moonshot AI$0.73 / 1M tokens$3.49 / 1M tokens262.1k
JSON modeLong contextStreamingTool calling
long Chinese documents, contract review1400-4400msCatalogOpenRouter if available

FAQ

Long-context models FAQ

Is a larger context window always better?

No. Larger context helps with big inputs, but cost, latency, retrieval design, and answer quality still matter.

すべてのモデル料金計算OpenAI互換クイックスタート