ToolsCompaniesAbout

AI Gateway

by TrueFoundry@truefoundryNo ratings yet

Single gateway for routing, governing, and monitoring LLM and MCP traffic

Visit truefoundry.com
Tool type
Enterprise AI gateway for LLMs and MCP tools
Made for
Platform, ML, and engineering teams
Runs on
SaaS gateway endpoint; VPC, on-prem, or air-gapped on Enterprise
Applications call the gateway endpoint over HTTP; the Playground and UI are browser-based.
Last checked
29 Oct 2026
3 product sources

What it does

TrueFoundry's AI Gateway sits between applications and model providers. It accepts LLM requests (chat, completion, embedding, and reranking) and MCP tool calls, then routes them to hosted models such as OpenAI, Claude, Gemini, Groq, and Mistral, or to self-hosted open-source models. Applications use one gateway endpoint and one gateway key, so models can be swapped through configuration rather than code changes.

Teams use it as a central control point for governance. It supports rate limits and token- or cost-based quotas per user, service, or endpoint, role-based access control, and guardrails such as PII filtering. Latency-based and weighted routing plus fallback chains handle provider failures. Policies can be managed in the UI or declared in YAML and applied through the CLI, which enables a GitOps workflow.

The gateway suits platform and ML teams running several models or agents across groups. The Developer plan is free for up to three users with 10K requests and tool calls per user per month. Pro is $25 per user per month, and Enterprise is a custom annual commitment. Model-provider token charges are billed separately by those providers.

What it can do

  • Unified API access to 1,600+ models, including OpenAI, Claude, Gemini, Groq, and Mistral
  • Observability: token usage, latency, error rates, and full request/response logs with metadata tagging
  • Quota and access control: per-user, per-service, or per-endpoint rate limits and cost- or token-based quotas with RBAC
  • Routing and fallbacks: latency-based, weighted load balancing, geo-aware routing, and automatic failover to secondary models
  • Self-hosted model serving for LLaMA, Mistral, and Falcon with vLLM, SGLang, KServe, and Triton compatibility
  • Built-in guardrails for PII filtering and toxicity detection, with custom and third-party guardrails on Enterprise

Works with

  • OpenAI
  • Anthropic Claude
  • Google Gemini
  • Groq
  • Mistral
  • AWS Bedrock
  • Azure
  • vLLM
  • SGLang
  • KServe

Before you choose

  • Model-provider token charges are separate and billed by OpenAI, Anthropic, Google, Azure, AWS, and similar providers; the gateway fee does not include them.
  • Third-party and custom guardrails (Bedrock, Azure, OpenAI, and others) are Enterprise-only and bill on the provider's usage; built-in guardrails are included on Pro and Enterprise.
  • Pro pricing uses pooled included requests per user with $20 per additional 100K overage; Agent Gateway and Agent Harness pricing is custom for now.

Listed under

Get started

Create an account, add a provider API key as a model account, then test in the Playground.

  1. Sign up at TrueFoundry and verify your email, then add a model account with your provider API key under AI Gateway > Models.
  2. Test models in the LLM Playground, then open the Code Snippet tab to copy the Gateway Base URL, API key, and Model ID.
  3. Create a Personal Access Token from the Access section for development, or a Virtual Account Token for applications, and point your SDK at the gateway base URL.
Sign up and add a model account

Plans and prices

Prices in USD

Gateway fees are billed by TrueFoundry; model-provider token charges are passed through separately. Pro includes 20K requests and tool calls per user per month, pooled across the team, with $20 per additional 100K overage. Trial length for Pro is not stated.

DeveloperFreeFree forever for up to 3 users.
  • Up to 3 users
  • 10K requests and tool calls per user per month
Pro$25/user/mo$25 per user per month; included requests pooled across the team.Overage $20 per additional 100K requests; see the pricing page rate table.
  • Unlimited users
  • 20K requests and tool calls per user per month
EnterpriseCustomCustom annual commitment; contact sales.
  • Custom log retention and audit exports
  • VPC, on-prem, and data residency

Source: truefoundry.com/pricing, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

Ratings and reviews

No ratings for AI Gateway yet.

Be the first to rate it

What’s changed

Free account

See what’s changed at AI Gateway

Launches, price moves and pivots, each with a date and a source. Free with an account.

Signals

Free account

See AI Gateway’s signals

Launch traction, hiring and release activity, tracked over time. Free with an account.

Data and security

Free account

See how AI Gateway handles your data

Training on your files, compliance, hosting and the models behind it. Free with an account.

Compare with similar tools

Free account

Compare AI Gateway side by side

Prices, platforms and features against similar tools, with sources. Free with an account.

Where this comes from

How we research
Company website
truefoundry.com: product, pricing and policy pages
YC directory mirror
Batch and company name
Social accounts
Accounts the company links, and its posts on X
Human review
Checked by a person
knowledge_base
manual_review

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.