Pricing Overview
Edgee takes a straightforward approach to pricing: start free, pay only for what you use, and never worry about markup on LLM provider costs. As an AI gateway that sits between your application and LLM providers, Edgee passes through provider pricing at cost while offering optional edge services like token compression, routing, and observability on top.
The core value proposition is simple. Edgee compresses prompts at the edge using intelligent token compression, reducing input tokens by up to 50% without losing semantic meaning. That means your actual LLM bills drop significantly, and the gateway itself charges no markup on the underlying provider rates. Enterprise pricing is available upon contact for organizations that need dedicated support, custom SLAs, or advanced governance features.
We find this model particularly appealing for teams that want to control AI costs without locking into a rigid per-seat or per-tier structure. You bring your own API keys (or use Edgee's keys for convenience), and Edgee handles the compression, routing, and cost governance layer.
Plan Comparison
Edgee structures its offering around a usage-based model with optional services rather than traditional fixed tiers. Here is how the key components break down:
| Component | What You Get | Cost Structure |
|---|---|---|
| AI Gateway | Route requests across 200+ models via a single OpenAI-compatible API | Provider pricing, no markup |
| Token Compression | Up to 50% input token reduction with semantic preservation | Included in gateway usage |
| Multi-Provider Routing | Normalize responses across OpenAI, Anthropic, Gemini, xAI, Mistral, and more | Included |
| Cost Governance | Tag requests by team/feature/project, spending alerts, anomaly detection | Included |
| Observability | Monitor latency, errors, usage, and cost per model/app/environment | Included |
| Edge Tools | Invoke shared tools or deploy private tools at the edge | Optional service |
| Edge Models | Run small, fast models at the edge for classification, redaction, enrichment | Optional service |
| Private Models | Deploy serverless open-source LLMs on demand through the same gateway API | Optional service |
| Bring Your Own Keys | Use your own provider API keys for billing control and custom models | Supported |
| Enterprise | Custom SLAs, dedicated support, advanced governance | Contact sales |
The free entry point lets you start routing requests immediately. Core features like token compression, multi-provider routing, cost governance, and observability come bundled with gateway usage. Advanced capabilities such as edge models, private model deployment, and custom edge tools are available as optional paid services.
Hidden Costs and Considerations
While Edgee's "no markup" positioning is attractive, there are several factors we think teams should evaluate before committing:
LLM provider costs are separate. Edgee does not bundle LLM usage into its pricing. You still pay OpenAI, Anthropic, or whichever provider you route through. Edgee's value comes from reducing those costs via token compression, not from offering a cheaper rate on the models themselves.
Optional services pricing is not published. Edge Tools, Edge Models, and Private Models are listed as optional services, but Edgee does not publish specific pricing for these add-ons on their website. Teams planning to use these advanced features will need to contact Edgee directly for quotes, which makes it harder to budget upfront.
Enterprise terms require a conversation. If you need custom SLAs, dedicated support, or advanced governance beyond what the standard offering provides, you will need to go through a sales process. There are no published enterprise tier rates.
Compression savings vary. The "up to 50% input token reduction" figure is a ceiling, not a guarantee. Actual savings depend on prompt structure, redundancy in your inputs, and use case. Teams running highly optimized, short prompts may see lower compression ratios than those with verbose, multi-turn agent workflows or large RAG contexts.
Vendor lock-in considerations. While Edgee uses an OpenAI-compatible API and supports bringing your own keys, routing all AI traffic through a single gateway creates a dependency. If Edgee experiences downtime or changes pricing terms, your entire AI stack is affected.
Open-source option exists. Edgee's gateway is open-source under the Apache-2.0 license (written in Rust), which means self-hosting is technically possible. However, self-hosted deployments would not include Edgee's managed edge network with 100+ global points of presence.
How Edgee Pricing Compares
Edgee occupies a unique position in the AI platforms space as a cost-optimization gateway rather than a direct LLM provider or a traditional SaaS tool. Here is how it stacks up against alternatives:
| Tool | Pricing Model | Starting Price | Key Difference |
|---|---|---|---|
| Edgee | Usage-based, no markup | $0.00 (free to start) | Gateway with token compression; you pay providers directly |
| Fusedash | Usage-based token packs | $0.00 (free tier) | Token packs at $5, $15, $25; bundled usage model |
| HypeScribe | Subscription tiers | $6.99/mo | Fixed tiers with transcription limits ($6.99-$12.99/mo) |
| Anthropic | Freemium + subscription | $0.00 (free tier) | Direct LLM provider; Pro at $20/mo, Team at $25/user/mo |
Edgee differentiates itself by not competing with LLM providers on model access. Instead, it reduces what you pay those providers. A team spending $10,000/month on OpenAI API calls could theoretically cut that to $5,000-$7,000/month through Edgee's token compression alone, while gaining multi-provider routing, cost governance dashboards, and observability at no markup.
Compared to Fusedash's token pack model ($5, $15, $25 tiers), Edgee avoids the complexity of pre-purchasing token bundles. You simply route traffic through the gateway and pay your LLM providers directly at their standard rates, minus whatever savings the compression engine delivers.
For teams already committed to a single provider like Anthropic (where Pro costs $20/month and Team runs $25/user/month), Edgee adds value as a complementary layer rather than a replacement. You can keep your Anthropic subscription while routing calls through Edgee to compress tokens and gain cross-provider routing flexibility.
The open-source nature of Edgee's core gateway (Apache-2.0, written in Rust) gives it a transparency advantage that most commercial gateways lack. Teams can inspect the compression logic, contribute to the project, or self-host if needed, which reduces long-term pricing risk.