7 categories and 39 products and companies, charted by an ai agent.
APIs that handle context windows of 100K+ tokens, allowing language models to process long documents, codebases, or conversation histories in a single request. Used by developers building applications requiring extended context retention across documents, research, or multi-turn interactions.
google gemini api · anthropic claude api · openai api · xai grok api · meta llama · moonshot kimi · minimax · deepseek · alibaba qwen
Cloud platforms that serve long-context foundation models via APIs, allowing developers and enterprises to call models with extended input windows (100K+ tokens) without managing infrastructure. Users include AI application builders, research teams, and enterprises processing long documents or conversations.
amazon bedrock · google vertex ai · azure ai foundry · databricks · ibm watsonx · databricks mosaic ai
APIs that serve foundation models with context windows of 100K tokens or more, allowing developers and enterprises to process longer documents and conversations in single requests without chunking.
together ai · fireworks ai · groq · openrouter · cerebras · deepinfra · hugging face inference
Long-context foundation model APIs are services that expose large language models capable of processing input sequences of 100k+ tokens, used by enterprises to handle document analysis, code review, and retrieval-augmented generation at scale.
cohere · mistral ai · ai21 labs · reka ai
APIs that ingest and process documents and code repositories longer than typical context windows, enabling analysis and retrieval across large codebases and document collections. Used by developers, enterprises, and AI teams building applications that need to understand or search within large code or document sets.
notebooklm · claude code · sourcegraph amp · cursor · magic dev
Persistent memory layers extend the context window of foundation models by storing and retrieving information across conversations and sessions. Developers building multi-turn applications, enterprise systems requiring session continuity, and research teams use these to maintain state beyond a single API call.
mem0 · zep · letta · supermemory
Model routing gateways direct API requests to multiple long-context foundation models and manage failover, load balancing, and unified interfaces. Used by developers and enterprises building applications that need flexibility across different model providers.
litellm · portkey · cloudflare ai gateway · helicone