Compresr provides an API for LLM context compression that reduces long context to the essentials a model needs without sacrificing quality. It functions as a drop-in solution for agents and RAG systems, helping reduce token costs and improve accuracy through context optimization. The founders come from EPFL research backgrounds. The company is part of Y Combinator's Winter 2026 batch.
Compresr
ActiveLLM context compression for better accuracy
Total raised
$500K
1 round
Stage
Seed
Jan 2026
Team
1-10
since 2026
Pricing
—
Founded
2026
Agent-ready
—
API for LLM context compression
Trims long context to the essentials a model needs
Drop-in integration for existing agent and RAG pipelines
Reduces token usage to cut inference costs
Aims to improve accuracy by removing noise from context
Quality-preserving compression rather than naive truncation
Built on research expertise from EPFL backgrounds
12/100
Early
MCP server
Public API
Webhooks
OAuth 2.0
SDKs
No public agent surfaces detected yet.
Jan 2026 Seed $500K ● Y Combinator
Capital network
$500K raised ·1 backer·10 network links
- Backers1
- Shared portfoliocompanies these backers also fund
- Extended networkfunds that co-invest alongside them
Databricks
The data + AI company
AI AgentsAI Infrastructure
Figure AI
General-purpose humanoid robots
AI InfrastructureAI Robotics
Upscale AI
Pure-play AI networking infrastructure
AI Developer ToolsAI Infrastructure
Dash0
AI-native observability platform built on OpenTelemetry
AI InfrastructureAI Data Engineering
Noma Security
End-to-end security for agentic AI
AI InfrastructureAI for Cyber Defense
Ineffable Intelligence
An AI research company building a superlearner to achieve superintelligence through reinforcement learning
Foundation ModelsAI Infrastructure
- How does Compresr differ from simple truncation?
- Rather than cutting off context arbitrarily, it reduces long context to the essential information a model needs, aiming to preserve quality while shrinking token count.
- Is it hard to integrate?
- It is offered as a drop-in API for agents and RAG systems, so it can be added to existing pipelines without re-architecting them.
- Does compression hurt accuracy?
- The product is designed to improve accuracy by removing irrelevant or noisy context, in addition to reducing token costs.
- What workloads benefit most?
- Long-context, high-volume use cases such as RAG and multi-step agents benefit most, since that is where token costs and context noise are highest.
Discussion
Sign in to join the discussion.
Sign inExplore more around Compresr
Contextual paths to related AI startups, deals and rankings.