One API. Every model. Total control.
Stop managing separate provider accounts. One endpoint gives you access to every major LLM with routing, cost controls, and observability included by default.
Your models stay up, even when providers don't
Connect to every major LLM through one endpoint. Merge routes each request automatically by cost, latency, or quality.
Every new model Merge supports is available to you automatically. When a provider goes down, failover is instant. No code changes required.
Set a budget. We'll enforce it.
Cap spend by project, team, or customer tier in real time. Cut token costs automatically with context compression.
Attribute every dollar to a model, provider, and team. Consolidate all provider invoices into one.
Control what reaches your models, and what doesn't
Stay in compliance with data loss prevention (DLP), prompt injection protection, and more. Filter sensitive data before it reaches any provider.
Enforce per-customer model and region policies. Log every routing decision, cost, and outcome in one place.
.avif)
Build Your Own Router
Define what "best" means for your product. Merge scores every available model against your weighted benchmarks and picks the winner on each request.
Routing presets
Start from balanced, code-heavy, reasoning-heavy, knowledge-heavy, or Instruction-following. Adjust weights or build from scratch.
Benchmark library
Choose from standard ML benchmarks (HumanEval, MMLU, GPQA, GSM8K), industry leaderboards (LMSYS Arena Elo, Aider, Contra), or add your own.
Scoring rules
Override or add your own benchmarks and scoring for any model via the dashboard, API, or JSON upload. The routing logic lives in your config, not in Merge's ML model.
Decision transparency
See which model was chosen, why it won, which rules fired, and where each score came from.
One line of code.
Deploy once.
No excuses.
Change the base URL to gateway.merge.dev/v1.
Every model, routing rule, and cost control is live from the first request.
One API key replaces every provider signup
Access all LLMs with one API. No separate accounts, no scattered credentials, no multiple invoices to manage.
Switch providers with a config change, not a refactor
Test any model head-to-head. Migrate away from a vendor without touching your application code.
Access every new model the day it ships
New providers and versions available automatically. No integration updates, no release delays.
See Gateway in action
“Thanks to Merge Gateway’s LLM routing functionality and spend cap rules, we’re able to save more than $10,000 per month on LLM spend without sacrificing performance.”
FAQ
*Bank or ACH gives unlimited zero-fee spend. The ~$40,000 credit card figure is derived from $100/month in fee coverage ($1,200 over 12 months) at standard credit card rates (approx 3%); fees beyond $100/month will be billed at standard credit card rates. Limited to 1 account per company, verified by email domain. Subject to verification and offer terms. See merge.dev/offer-terms
Your AI roadmap is ambitious. Your infrastructure should match.
Start for free. Zero fees for 12 months when you sign up by June 30.
