Best for: LiteLLM lists virtual keys, budgets, RPM/TPM limits, spend by key, user, team and organisation, tag-based spend tracking and audit logs.
Categories › AI Gateway Software
Best LLM Cost Control and Budget Tools for AI Gateways 2026
LiteLLM is the pick for budgets and RPM/TPM limits per key, user and team on a gateway you run yourself. Vercel AI Gateway is the pick for spend budgets at team, project, key or user scope with no markup on tokens, and TrueFoundry is the pick for budget rules on a $25-per-user plan. Azure API Management is the pick for per-consumer token quotas on LLM APIs, and OpenRouter is the pick for spend controls on a hosted multi-model API.
Updated
Compare the deciding features
| Product | Documented buyer fit |
|---|---|
| LiteLLM | LiteLLM lists virtual keys, budgets, RPM/TPM limits, spend by key, user, team and organisation, tag-based spend tracking and audit logs.[1] [2] [3] |
| Vercel AI Gateway | Vercel AI Gateway budgets cap spend at team, project, API key or user scope.[4] [5] [6] |
| TrueFoundry | TrueFoundry Pro costs $25 per user a month with unlimited seats, SSO and custom roles; the Enterprise plan runs on an annual commitment.[7] |
| Portkey | Portkey’s Enterprise plan lists SSO, granular budget and rate limits, private cloud and VPC hosting, and compliance covering SOC 2 Type 2, GDPR and HIPAA.[8] |
| OpenRouter | OpenRouter budgets and spend controls and auto-routing are included from the Standard plan.[9] |
| Kong AI Gateway | Kong AI Gateway model access and token budgets can be scoped by team or department, and usage analytics cover requests, tokens, errors and latency.[10] [11] [12] [13] |
| Cloudflare AI Gateway | Cloudflare lists AI Gateway's core features as free, and they include dashboard analytics, caching and rate limiting.[14] [15] [16] |
| Azure API Management | Azure API Management token limit policy enforces a tokens-per-minute limit or a quota per hour, day, week, month or year for each API consumer.[17] |
| Eden AI | Eden AI gives one API to about 1,000 models from 80+ providers, with a 5.5% platform fee applied when you buy credits.[18] [19] |
| Requesty | Requesty offers 600+ models through one API with routing and failover, spend limits, governance and observability of every request.[20] |
Ranked shortlist
Best for: Vercel AI Gateway budgets cap spend at team, project, API key or user scope.
Best for: TrueFoundry Pro costs $25 per user a month with unlimited seats, SSO and custom roles;
Best for: Portkey’s Enterprise plan lists SSO, granular budget and rate limits, private cloud and VPC hosting, and compliance covering SOC 2 Type 2, GDPR and HIPAA.
Best for: OpenRouter budgets and spend controls and auto-routing are included from the Standard plan.
Best for: Kong AI Gateway model access and token budgets can be scoped by team or department, and usage analytics cover requests, tokens, errors and latency.
Best for: Cloudflare lists AI Gateway's core features as free, and they include dashboard analytics, caching and rate limiting.
Best for: Azure API Management token limit policy enforces a tokens-per-minute limit or a quota per hour, day, week, month or year for each API consumer.
Best for: Eden AI gives one API to about 1,000 models from 80+ providers, with a 5.5% platform fee applied when you buy credits.
Best for: Requesty offers 600+ models through one API with routing and failover, spend limits, governance and observability of every request.
Evidence and sources
- LiteLLM official page or repositoryRetrieved 2026-11-10T01:13:00Z
Virtual keys, budgets, RPM/TPM limits Spend by key, user, team & org Tag-based spend tracking Audit logs on every request
- LiteLLM official page or repositoryRetrieved 2026-11-10T01:13:02Z
The owner's tpm_limit and rpm_limit whenever they are set
- LiteLLM official page or repositoryRetrieved 2026-11-10T01:13:02Z
Load-balance across multiple deployments (e.g. Azure/OpenAI)
- Vercel AI Gateway official page or repositoryRetrieved 2026-11-10T01:14:52Z
A budget caps AI Gateway spend at one of four scopes: Team : everything your team spends, across every project and key. Project : what one project's deployments spend through their OIDC tokens, plus what API keys scoped to that project spend. API key : what one API key spends. User : what one team member spends, across every API key attributed to them.
- Vercel AI Gateway official page or repositoryRetrieved 2026-11-10T01:12:12Z
AI Gateway charges no markup and no platform fee on tokens. You pay the provider's list price on a pay-as-you-go basis.
- Vercel AI Gateway official page or repositoryRetrieved 2026-11-10T01:12:11Z
One API key, 400+ models.
- TrueFoundry official page or repositoryRetrieved 2026-11-10T01:12:12Z
Developer $0 free forever For small teams prototyping AI workflows
- Portkey official page or repositoryRetrieved 2026-11-10T01:12:10Z
Role-Based Access Control, SSO, Granular Budget & Rate Limits
- OpenRouter official page or repositoryRetrieved 2026-11-10T01:12:10Z
Budgets & Spend Controls Free No Standard Yes Business Yes Enterprise Yes
- Kong AI Gateway official page or repositoryRetrieved 2026-11-10T01:12:10Z
Scope model access and token budgets by team or department instead of managing permissions per person
- Kong AI Gateway official page or repositoryRetrieved 2026-11-10T01:13:02Z
AI Management RECOMMENDED Plus From $25/month plus usage, billed monthly Start governing smaller LLM, MCP, and A2A deployments.
- Kong AI Gateway official page or repositoryRetrieved 2026-11-10T01:13:37Z
AI Gateway Enterprise: This plugin is only available as part of our AI Gateway Enterprise offering.
- Kong AI Gateway official page or repositoryRetrieved 2026-11-10T01:13:37Z
AI Gateway Enterprise: This plugin is only available as part of our AI Gateway Enterprise offering.
- Cloudflare AI Gateway official page or repositoryRetrieved 2026-11-10T01:12:10Z
AI Gateway's core features available today are offered for free, and all it takes is a Cloudflare account and one line of code to get started
- Cloudflare AI Gateway official page or repositoryRetrieved 2026-11-10T01:13:00Z
For example, you can limit your application to 100 requests per 60 seconds. You can also select if you would like a fixed or sliding rate limiting technique.
- Cloudflare AI Gateway official page or repositoryRetrieved 2026-11-10T01:13:00Z
AI Gateway can cache responses from your AI model providers, serving them directly from Cloudflare's cache for identical requests.
- Azure API Management official page or repositoryRetrieved 2026-11-10T01:12:10Z
Configure a token limit policy on your LLM APIs to manage and enforce limits per API consumer based on the usage of AI service tokens.
- Eden AI official page or repositoryRetrieved 2026-11-10T01:12:10Z
Access to ~1,000 models from 80+ providers with one API
- Eden AI official page or repositoryRetrieved 2026-11-10T01:13:00Z
Attribute every call to a team, client or product, with budgets, roles and model access per key.
- Requesty official page or repositoryRetrieved 2026-11-10T01:12:11Z
Access 600+ models through one API with routing and failover, spend limits, governance and full observability of every request.
1. LiteLLM: best for budgets and rate limits per key, user and team on a gateway you run
LiteLLM lists virtual keys, budgets, RPM/TPM limits, spend by key, user, team and organisation, tag-based spend tracking and audit logs. Key owners can carry tpm_limit and rpm_limit values. Routing load-balances across multiple deployments (for example Azure and OpenAI) with cooldowns and fallbacks.
Sources: LiteLLM source · LiteLLM source · LiteLLM source
2. Vercel AI Gateway: best for spend budgets by team, project, key or user with zero token markup
Vercel AI Gateway budgets cap spend at team, project, API key or user scope. It charges no markup and no platform fee on tokens, and includes $5 a month of credit. Vercel AI Gateway offers one API key for 400+ models and automatic fallbacks during provider outages.
Sources: Vercel AI Gateway source · Vercel AI Gateway source · Vercel AI Gateway source
3. TrueFoundry: best for budget rules on a per-seat plan
TrueFoundry Pro costs $25 per user a month with unlimited seats, SSO and custom roles; the Enterprise plan runs on an annual commitment. Budget rules are capped at 1 on Developer and 10 on Pro, and unlimited on Enterprise. Pro includes 20K requests and tool calls per user a month (then $20 per additional 100K requests), built-in guardrails, budget limiting and 30-day log retention.
Sources: TrueFoundry source
4. Portkey: best for budget and rate-limit governance on the Enterprise plan
Portkey’s Enterprise plan lists SSO, granular budget and rate limits, private cloud and VPC hosting, and compliance covering SOC 2 Type 2, GDPR and HIPAA. The Production plan costs $49 a month and includes 100k recorded logs, with $9 per additional 100k requests, 30 days of log retention and 90 days of metrics.
Sources: Portkey source
5. OpenRouter: best for spend controls on a hosted multi-model API
OpenRouter budgets and spend controls and auto-routing are included from the Standard plan. Inference is billed at the provider's list price; the fee is 5.5% on Standard and 8% on Business, charged when credits are bought, never on individual requests. Standard and Business include $25,000 of list-price inference a month through your own provider keys with no OpenRouter fee.
Sources: OpenRouter source
6. Kong AI Gateway: best for token budgets by team
Kong AI Gateway model access and token budgets can be scoped by team or department, and usage analytics cover requests, tokens, errors and latency. The AI Management Plus plan starts at $25 a month plus usage, billed monthly; the Enterprise plan is billed annually. Some AI Gateway plugins are available only in the AI Gateway Enterprise offering.
Sources: Kong AI Gateway source · Kong AI Gateway source · Kong AI Gateway source · Kong AI Gateway source
7. Cloudflare AI Gateway: best for request rate limits on Cloudflare
Cloudflare lists AI Gateway's core features as free, and they include dashboard analytics, caching and rate limiting. Rate limits are set as a number of requests in a time frame, for example 100 requests per 60 seconds, with a fixed or sliding window. Caching serves identical requests from Cloudflare's cache instead of calling the model provider.
Sources: Cloudflare AI Gateway source · Cloudflare AI Gateway source · Cloudflare AI Gateway source
8. Azure API Management: best for per-consumer token quotas on LLM APIs
Azure API Management token limit policy enforces a tokens-per-minute limit or a quota per hour, day, week, month or year for each API consumer. Azure API Management's AI gateway features cover load balancing across AI endpoints, logging of AI interactions and token quotas across applications.
Sources: Azure API Management source
9. Eden AI: best for per-key budgets
Eden AI gives one API to about 1,000 models from 80+ providers, with a 5.5% platform fee applied when you buy credits. It lets you attribute every call to a team, client or product, with budgets, roles and model access per key.
Sources: Eden AI source · Eden AI source
10. Requesty: best for spend limits on a hosted multi-model router
Requesty offers 600+ models through one API with routing and failover, spend limits, governance and observability of every request.
Sources: Requesty source
How it works
An AI gateway controls LLM spend by identifying who is calling (a virtual key, team or user), counting tokens or dollars per request, and blocking or alerting when a limit is reached. Rate limits cap request or token volume over a time window; budgets cap cumulative spend. Gateways differ in scope (key, user, team, project), in the unit (spend or tokens) and in which plan includes the control.
How to choose
Pick the scope that matches how your teams are billed: per team, per project, per key or per user. Check whether budgets and rate limits are in the plan you will buy. Portkey lists granular budget and rate limits on Enterprise, and Kong lists some AI Gateway plugins as Enterprise-only. Check any platform fee on credits (5.5% on OpenRouter Standard, 5.5% on Eden AI) against the control you gain.
Frequently asked questions
What software sits between our apps and model providers to set budgets per team?
LiteLLM, Vercel AI Gateway, TrueFoundry, Kong AI Gateway and Portkey Enterprise all set budgets by team or key. LiteLLM lists virtual keys, budgets and RPM/TPM limits with spend by key, user, team and organisation.
How do I set a budget per API key for LLM calls?
Vercel AI Gateway budgets cap an API key, a project, a team or a user. LiteLLM virtual keys carry budgets and tpm and rpm limits.
Can I rate limit LLM requests at the gateway?
Cloudflare AI Gateway sets request limits such as 100 requests per 60 seconds with fixed or sliding windows, and Azure API Management enforces tokens-per-minute limits or quotas per hour to year.
Which gateway gives spend visibility without a markup on tokens?
Vercel AI Gateway charges no markup and no platform fee on tokens and logs spend and model usage. OpenRouter bills at provider list price with a fee on credit purchases.
How do I stop LLM costs spiking across five product teams?
Give each team its own key or scope with a budget: LiteLLM lists spend by team, TrueFoundry lists budget rules, and Kong AI Gateway scopes token budgets by team or department.
Methodology
Every price, limit, licence and feature comes from the vendor's or project's own pages, repository files and documentation, opened on 10 November 2026 (Europe/Helsinki), with the exact URL and a verbatim excerpt recorded for each fact. Prices are shown in the currency the vendor prints, with the billing period named. Picks follow documented fit for the buying situation and published facts, not paid placement or review-site scores. Vendor statements about compliance are quoted as the vendor states them. Vendor families are counted once.
How this comparison is made
Ranked on documented capabilities and verified pricing from official vendor sources. Every fact links to its source and check date, and the criteria are specific to this buying situation.
How we assessed this shortlist
Every price, limit, licence and feature comes from the vendor's or project's own pages, repository files and documentation, opened on 10 November 2026 (Europe/Helsinki), with the exact URL and a verbatim excerpt recorded for each fact. Prices are shown in the currency the vendor prints, with the billing period named. Picks follow documented fit for the buying situation and published facts, not paid placement or review-site scores. Vendor statements about compliance are quoted as the vendor states them. Vendor families are counted once.
Want your software in this comparison? A vendor profile listing costs $9.90 via Stripe and goes live within 24 hours of payment. Comparison-list placements cost a fee, and links are dofollow. Factual corrections are free.
Submit your software →Spot an error or a missing product?
Send a correction with a source link and we review it against the vendor's own documentation.
Suggest a correction