Compresr API
A query-aware compression API that drops the context tokens a model will not use before an LLM call, offered through Python and TypeScript SDKs, a hosted HTTP API or on-prem inside the customer's VPC.
Find alternatives to Compresr API
What it does
- Cloud and AI cost management
- Model inference
Features 5
- Question-aware compressionTakes a query and a context, then returns only the tokens that move the answer.
- RAG compressionCompresses retrieved chunks before the model sees them, so paid tokens carry the information.
- Tool output compressionCompresses agent tool results, using the tool call's intent as the query.
- LangChain middlewaresThree LangChain middlewares: tool output, history summarization and prompt budget.
- On-prem deploymentThe compression engine can run inside the customer's own network.
Sources
- Pricing
-
Pricing page: "Every plan is scoped to your workload. Book a call and we'll size it with you."
- Description
-
Official documentation · 11 Nov 2026
Product page: "Compresr is a context-compression API for LLM developers."
- What it does
-
Official documentation · 11 Nov 2026
Product page: "Compresr is a context-compression API for LLM developers."
- Sold within
-
Sold on its own with published plans.
- Name
-
Official documentation · 11 Nov 2026
Name as published on the product page.
- URL
-
Official documentation · 11 Nov 2026
Product page, read 2026-11-11.
- Deployment
-
Official documentation · 11 Nov 2026
Product page: "Cloud API or on-prem inside your VPC"
- Interfaces
-
Official documentation · 11 Nov 2026
Product page: "Cloud API or on-prem inside your VPC"
Open Compresr in the directory
Something wrong here? Send a correction — quote this product id: compresr-api.