Personal
$0
- Free for personal use
- Runs through the dense CLI
Respan Gateway: up to 55% off tokens on Gemini, GPT, Claude and 1,000+ more models. One API key, 0% fee.
Condense is a drop-in proxy that manages and compresses the context an application or agent sends to the LLM on every call. Teams change one base URL in their Anthropic or OpenAI setup and keep their own provider key, so the provider still bills them directly, just for fewer tokens. Condense reports up to 72% fewer tokens on real traffic.
Condense trains its own compression models. Helene 1.1 strips low-value tokens from each request and is the proxy's default, while Adeline 1 handles long agent sessions by rewriting older, finished work. Beyond Anthropic and OpenAI, it has routes for Amazon Bedrock and Vertex AI, and developers can run Claude Code, Codex, or OpenCode through it with the dense CLI.
Core capabilities this platform advertises.
What this tool does well, and the limitations to keep in mind.
Pros
Cons
What's included in each plan, and how the tiers compare.
$0
15% of savings
Usage-based
$0.40 to $0.90 per 1M tokens removed
Usage-based
Developers and teams running agents on Anthropic or OpenAI who want lower token bills without changing their code.
Condense cuts the context going into each LLM call, and Respan shows what comes back. Route, observe, and evaluate every LLM call to confirm that compressed requests still produce good output.
Respan lets you trace LLM and agent calls across any model or framework, A/B test prompts on production traffic, and route requests across 1,000+ models through one gateway.
Top companies in LLM Gateways you can use instead of Condense.
Respan
One API for 1,000+ models
OpenRouter
One OpenAI-compatible API for 500+ models from 80+ providers
Cloudflare AI Gateway
Edge-deployed AI gateway
Vercel AI Gateway
One API key for hundreds of models
LiteLLM
OpenAI-compatible proxy for 140+ providers
Portkey
Unified API across major providers
Bifrost
OpenAI-compatible API for 1,000+ models from 20+ providers
Helicone
LLM observability and monitoring
Stainless
SDK generation
Unify
Martian
Intelligent model routing based on prompt type
Kong AI Gateway
AI traffic management
Apigee AI Gateway
Google Cloud AI traffic management
Requesty
400+ model access
The Token Company
Token compression
TrueFoundry
AI Gateway with routing, fallbacks, rate limits, and budgets
OrcaRouter
orcarouter/auto adaptive routing per request
InWorld AI
Conditional routing with CEL expressions on request metadata
Orq.ai
AI gateway for 500+ models with fallbacks, retries, and load balancing
Side-by-side comparisons with other tools in this category.
Companies from adjacent layers in the AI stack that work well with Condense.