Guides
Budgets vs Rate Limits: Which Control to Reach For
Both return 429, but they solve different problems. Here is a decision guide for when to use a budget cap, when to use a rate limit, and why you almost always want both.
Nemo Team
7 minPosts
Latest first
Budgets vs Rate Limits: Which Control to Reach For
Uptime99.998%
Cost Saving-60.4%
Cost-honest controlSafety Guardrails
Active & Safe
Nemo Routernemorouter.aiHandling 429 and 402 Errors From an LLM Gateway
Tenant ScopeMulti-Tenancy
API SecurityGuardrails
AccountingBudgets
Row-Level Security ActiveZero-Latency gateway
SOC2 ReadyNemo Routernemorouter.aiGuides
Handling 429 and 402 Errors From an LLM Gateway
A 429 and a 402 mean different things and need different client logic. Here is how to handle rate limits, budget caps, and out-of-credits responses gracefully — with backoff, not blind retries.
Nemo Team
8 minrate-limiting-rpm-tpm.ts
NEMO GATEWAY// Title: RPM and TPM Rate Limiting Per Key, Team, and Org
// 1. Initialize custom gateway router
const nemo = new NemoRouter({
apiKey: "sk-nemo-prod-...",
fallback: true,// Safe active failovers
});
// 2. Direct request routing
await nemo.chat.completions.create({ model: "claude" });
API ActivePort 3001
200 OK (240ms)Nemo Routernemorouter.aiEngineering
RPM and TPM Rate Limiting Per Key, Team, and Org
Rate limits cap velocity, not total spend — and they're a security boundary, not a knob. Here is how RPM/TPM limits work per key, team, and org, and why the caller can never override them.
Nemo Team
8 minThe Four Ceilings Every LLM Request Passes
Legacy Proxy
Platform Fee
4% Markup
Timeout rate: 529 errors
Nemo Router
Platform Fee
0% Tier 3
Failover: 0ms fallback
Compare vs standard proxiesActive Fallbacks
Nemo Routernemorouter.aiGuides
The Four Ceilings Every LLM Request Passes
Before an LLM call leaves the gateway it clears four independent limits: credit balance, budget cap, rate limit, and guardrails. Here is what each one checks and why they are separate.
Nemo Team
8 min