Metorial vs TrueFoundry AI Gateway

TrueFoundry AI Gateway (MCP Gateway) and Metorial both put a governed layer between AI clients and the tools your company uses.

Where TrueFoundry AI Gateway wins: One gateway for both LLM traffic and MCP tool calls, with per-user/team cost tracking and budget limits alongside tool governance.

Metorial is the stronger fit if you want self-hosting, Git-synced skills and ChatGPT support.

Every TrueFoundry AI Gateway fact below links to TrueFoundry AI Gateway's own page. Anything they don't publish is marked "Not published", not "No".

Where TrueFoundry AI Gateway wins

  • One gateway for both LLM traffic and MCP tool calls, with per-user/team cost tracking and budget limits alongside tool governance.source
  • Built-in pre-tool and post-tool guardrails on MCP calls, including PII redaction, prompt-injection detection and Cedar/OPA permission checks.source
  • Enterprise deployment into customer VPC, on-prem or air-gapped Kubernetes, with SOC 2 Type 2 and HIPAA claims.source

Key differences

Each cell links to the page it comes from.

FeatureMetorialTrueFoundry AI Gateway
Source licenseFSL-1.1-ALv2 engine (Apache 2.0 after two years); Apache-2.0 integration catalogsourceThe engine (metorial-platform) is source-available under FSL and converts to Apache 2.0 two years after each release. The integration catalog (metorial/metorial) is Apache-2.0.Not publishedVendor open-sources other projects (TrueForge agent harness, KubeElasti, CruiseKube).
Self-hostYessourcePartialsourceEnterprise only.
Integration catalog1,000+sourceNot publishedCurated registry of prebuilt servers (Slack, Confluence, Sentry, Datadog named); no count published.
One URL for all toolsYessourceYessourceVia Virtual MCP Servers; Developer plan limited to 1 virtual MCP.
Per-user OAuthYessourceYessource
SSO (SAML or OIDC)YessourcePartialsourcePricing lists SSO & custom roles on Pro and above; not on free Developer plan.
SCIM provisioningYessourceYessourcePricing FAQ ties SCIM to Enterprise.
Role or group-based tool accessYessourceYessourceRBAC is per MCP server; tool-level curation via virtual servers.
Audit log per personYessourceYessourceAccess resolved to user, team or virtual account identity.
Create and edit skillsYessourceYessource
Git sync for skillsYessourceNot publishedCLI upload and declarative tfy apply via CI/CD described; no Git sync.
Own chat app requiredNosourceNosource
Prompt-injection detectionYessourceYessource
EU data regionYessourceYessourceControl plane primary in eu-west-1.
SOC 2 Type IIYessourceYessourceListed under SOC 2 Type II.
LLM gateway or model routingNoYessource
Starting price$0 (Dev); Scale from $250/monthsource$25 per user/month (Pro); $0 Developer up to 3 userssource

Which AI clients each works in

FeatureMetorialTrueFoundry AI Gateway
Claude (web and desktop)YessourceNot publishedClaude Desktop/claude.ai not named in IDE connection docs.
Claude CodeYessourceYessource
ChatGPTYessourceNot published
CursorYessourceYessource
CodexYessourceNot published
GitHub CopilotYessourcePartialsourceVS Code documented; Copilot not named explicitly.
Custom agents (SDK or API)YessourceYessource

Full checklist

Every row we track, with what it means. "Not published" means the vendor's own pages don't say.

Connect

FeatureMetorialTrueFoundry AI Gateway
Integration catalogHow many ready-made integrations or MCP servers the vendor publishes.1,000+sourceNot publishedCurated registry of prebuilt servers (Slack, Confluence, Sentry, Datadog named); no count published.
Verified or official serversThe catalog includes servers the vendor tests or maintains, not only community listings.YessourcePartialsourceCurated, but no explicit verification program described.
Bring your own MCP serverYou can add an MCP server you built or host yourself.YessourceYessourceAlso OpenAPI-to-MCP conversion.
Remote (HTTP) MCP serversConnects to MCP servers hosted elsewhere over HTTP.YessourceYessource
Local, stdio or Docker serversRuns MCP servers that ship as local processes or containers.YessourcePartialsourceGateway runs stdio servers as processes (npx/uvx); docs say recommended for local development in IDEs only.
One URL for all toolsA user adds one endpoint to their AI client and reaches every approved tool through it.YessourceYessourceVia Virtual MCP Servers; Developer plan limited to 1 virtual MCP.
Version pinningYou can lock an integration to a known version.YessourceNot published
Schema change monitoringChanges to tool definitions are detected and flagged.YessourceNot published

AI clients

FeatureMetorialTrueFoundry AI Gateway
Claude (web and desktop)Documented to work in the Claude apps.YessourceNot publishedClaude Desktop/claude.ai not named in IDE connection docs.
Claude CodeDocumented to work in Claude Code.YessourceYessource
ChatGPTDocumented to work in ChatGPT.YessourceNot published
CursorDocumented to work in Cursor.YessourceYessource
CodexDocumented to work in Codex.YessourceNot published
GitHub CopilotDocumented to work in GitHub Copilot.YessourcePartialsourceVS Code documented; Copilot not named explicitly.
Custom agents (SDK or API)Your own agents can use the tools through an SDK or API.YessourceYessource
Own chat app requiredYes means people must work inside the vendor’s own app rather than the AI client they already use.NosourceNosource

Identity and access

FeatureMetorialTrueFoundry AI Gateway
Per-user OAuthTool calls run with each person’s own account, not a shared key.YessourceYessource
Managed OAuth appsThe vendor provides OAuth apps so you don’t register your own.YessourceNot published
Bring your own OAuth appYou can use OAuth apps registered under your own company.YessourceYessourceAdmin supplies OAuth client credentials per server.
SSO (SAML or OIDC)People sign in through your identity provider.YessourcePartialsourcePricing lists SSO & custom roles on Pro and above; not on free Developer plan.
SCIM provisioningUsers and groups sync automatically from your identity provider.YessourceYessourcePricing FAQ ties SCIM to Enterprise.
Role or group-based tool accessAccess to tools follows roles or groups.YessourceYessourceRBAC is per MCP server; tool-level curation via virtual servers.
Tool-level allow and denyIndividual tools can be allowed or blocked, not only whole servers.YessourceYessource
Read-only modesAn integration can be limited to read-only tools.YessourcePartialsourceAnnotation hint only, not an enforced read-only mode.

Visibility and security

FeatureMetorialTrueFoundry AI Gateway
Audit log per personEvery call is recorded and tied to the person or agent who made it.YessourceYessourceAccess resolved to user, team or virtual account identity.
Call tracingTool arguments and results are recorded for debugging and review.YessourceYessource
Log export or SIEMLogs can be exported or streamed to your security tools.YessourceYessourceOTel trace export; audit exports listed as Enterprise on pricing.
Data redaction or DLPSensitive data in tool traffic can be detected, masked or blocked.YessourceYessource
Prompt-injection detectionTool traffic is inspected for prompt-injection attacks.YessourceYessource
Tool-poisoning protectionMalicious or changed tool descriptions are detected.YesThrough Protoguard.PartialsourceNo explicit tool-poisoning feature; post-tool guardrails can inspect outputs.

Skills

FeatureMetorialTrueFoundry AI Gateway
Create and edit skillsTeams can write and edit shared skills in the product.YessourceYessource
Git sync for skillsSkills sync to or from a Git repository.YessourceNot publishedCLI upload and declarative tfy apply via CI/CD described; no Git sync.
Skills delivered by roleSkills reach people based on their role or group.YessourcePartialsourceAccess control mentioned; role/team assignment not detailed.
Skills over MCPSkills are served to AI clients through MCP.Not publishedNot published

Deployment

FeatureMetorialTrueFoundry AI Gateway
Source licenseThe license the product’s code is published under, if any.FSL-1.1-ALv2 engine (Apache 2.0 after two years); Apache-2.0 integration catalogsourceThe engine (metorial-platform) is source-available under FSL and converts to Apache 2.0 two years after each release. The integration catalog (metorial/metorial) is Apache-2.0.Not publishedVendor open-sources other projects (TrueForge agent harness, KubeElasti, CruiseKube).
Self-hostYou can run it on your own infrastructure.YessourcePartialsourceEnterprise only.
VPC or on-premCan be deployed in your own cloud account or data center.YessourcePartialsourceEnterprise plan per pricing page.
Air-gappedCan run with no internet connection.YesPartialsourceEnterprise plan.
EU data regionData can be hosted in the EU.YessourceYessourceControl plane primary in eu-west-1.

Compliance

FeatureMetorialTrueFoundry AI Gateway
SOC 2 Type IIHolds a SOC 2 Type II report.YessourceYessourceListed under SOC 2 Type II.
ISO 27001Holds ISO/IEC 27001 certification.Not publishedNot published
HIPAASupports HIPAA workloads, usually with a BAA.YesSigns BAAs.Yessource
GDPR and DPAStates GDPR compliance and offers a data processing agreement.YessourceA data processing agreement is published at /legal/data-processing-agreement.YessourceCustomer-facing DPA availability not explicitly stated; DPAs signed with sub-processors.

Pricing

FeatureMetorialTrueFoundry AI Gateway
Public pricingPrices are published on the vendor’s site.YessourceYessource
Free tierA free plan or free edition exists.YessourceYessourceUp to 3 users, SaaS only.
Starting priceThe lowest published paid price, as listed.$0 (Dev); Scale from $250/monthsource$25 per user/month (Pro); $0 Developer up to 3 userssource
Pricing modelHow the vendor charges: per seat, usage, platform fee, or a mix.Flat monthly plans with paid add-ons (extra seats, tool calls)sourcePer user/month plus request overage ($20 per additional 100K requests); Enterprise custom annualsource

Model layer

FeatureMetorialTrueFoundry AI Gateway
LLM gateway or model routingRoutes model traffic, not only tool traffic.NoYessource
Cost tracking per personShows AI spend per person or team.NoYessource

Which should you pick?

SituationPickWhy
Rolling AI out to every employee in the clients they already useEitherBoth connect to existing AI clients. Compare the client list below.
You must self-hostMetorialMetorial documents self-hosting; TrueFoundry AI Gateway's own pages don't state a self-hosted option.
You need an LLM gateway or model routing todayTrueFoundry AI GatewayTrueFoundry AI Gateway documents an LLM gateway. Metorial doesn't ship one yet.
Teams should share skills, not only toolsEitherBoth document shared skills; compare the Skills rows below.
You are building integrations into your own productEitherBoth offer an SDK or API for your own agents.

Pricing

Published prices only. Nothing is estimated.

Metorial

  • Metorial Dev: $0 monthsource
  • Metorial Scale: $250 monthsource
  • Metorial for Enterprise: Contact salessource

TrueFoundry AI Gateway

  • Developer: $0 free forever up to 3 userssource
  • Pro: $25 per user / month per user/monthsource
  • Enterprise: Custom annual commitment annual contractsource
  • Pro overage: + $20 per additional 100K requests
  • Model provider costs are pass-through, billed separately
  • Agent Gateway / Agent Harness pricing custom

What each Metorial plan includes

From the Metorial pricing page.

IncludedMetorial DevMetorial ScaleMetorial for Enterprise
Tool calls per month500,0002,500,000Custom
Callbacks10,000100,000Custom
Workforce seats220Custom
Provider integrations10UnlimitedUnlimited
Log retention30 days90 daysCustom
Production accessNot includedIncludedIncluded

On every plan

  • Slack and email support
  • Auth Connections: Unlimited
  • Sessions: Unlimited
  • Concurrent Connections: Unlimited
  • Developer API Access
  • Custom Integrations
  • Remote MCP Servers
  • Managed MCP Servers

TrueFoundry AI Gateway limits

As published by the vendor.

  • Developer plan users: Up to 3 userssource
  • Developer plan MCP servers: 5 MCP servers, 1 virtual MCPsource
  • Developer requests: 10K requests and tool calls per user per monthsource
  • Pro requests: 20K requests and tool calls per user per month; +$20 per additional 100Ksource
  • Log retention: 7 days (Developer), 30 days (Pro), custom (Enterprise)source
  • Tool description override: up to 20,000 characterssource

Moving from TrueFoundry AI Gateway to Metorial

  • Remote MCP servers you use through TrueFoundry AI Gateway can be added to Metorial as they are.
  • OAuth apps registered under your company can be reused, so people sign in the way they do today.
  • Sign-in moves to your identity provider through SAML; groups decide who gets which tools.
  • Skills, policies and audit history in TrueFoundry AI Gateway don't transfer automatically. Talk to us about moving them.

Book a demo to plan the move with us.

Frequently asked questions

Common questions about Metorial and TrueFoundry AI Gateway.

  1. Is TrueFoundry AI Gateway open source?
    TrueFoundry AI Gateway's own pages don't state a source license.
  2. Yes, TrueFoundry AI Gateway documents a self-hosted option. (with limits)
  3. TrueFoundry AI Gateway's own pages don't document ChatGPT support.
  4. Metorial's Dev plan is free and Scale starts at $250 a month. TrueFoundry AI Gateway publishes Developer at $0 free forever up to 3 users and Pro at $25 per user / month per user/month. The right answer depends on team size; see the pricing section.
  5. Yes. Metorial is SOC 2 Type 2 and GDPR compliant, with EU and US regions.
  6. Metorial's engine is published under FSL-1.1-ALv2 engine (Apache 2.0 after two years); Apache-2.0 integration catalog. You can self-host it or use the hosted platform.

Checked October 17, 2026. 38 TrueFoundry AI Gateway facts are sourced, each linked in its row. Spotted something wrong? Tell us and we will fix it. All product names are trademarks of their owners.

Try Metorial next to TrueFoundry AI Gateway

Start free, self-host from GitHub, or talk to us about your rollout.