Integrations: AI model providers

Argent with Cloudflare AI Gateway

Already route AI traffic through Cloudflare AI Gateway? Keep it. Argent adds the agent and the person behind each call, your approved models, redaction, budgets and a signed record.

01 / INTEGRATIONS: AI MODEL PROVIDERS

What Argent does with Cloudflare AI Gateway

Cloudflare AI Gateway routes and manages model traffic. Argent governs it: who is calling, what they may send and what it costs.

  • Built in: Cloudflare AI Gateway's connection settings ship with Argent, so agents reach it through Argent by changing their base URL and key.
  • Agent and owner on every call: each request is tied to a registered agent and its human owner.
  • Approved models: the model in each request is checked against your approved list and access rules before it reaches Cloudflare AI Gateway.
  • Sensitive data: redacted before the request leaves your network.
  • Budgets and record: every call priced, held against a budget and signed.
02 / INTEGRATIONS: AI MODEL PROVIDERS

How it connects

Agents point at Argent, and Argent forwards to your Cloudflare AI Gateway.

  1. Add Cloudflare AI Gateway as an endpoint on the AI Providers page. Its key goes to your vault and Argent keeps only the reference.
  2. Register each agent with an owner and give it its Argent key.
  3. Point agents at Argent. Their base URL becomes the gateway route; Argent attaches the Cloudflare AI Gateway key on the way out.

Your vault can be HashiCorp Vault or OpenBao, AWS Secrets Manager, CyberArk Conjur or Infisical.

03 / INTEGRATIONS: AI MODEL PROVIDERS

What you configure

Everything below is set once in the Argent console and applies to every agent that calls Cloudflare AI Gateway through the gateway.

SettingWhereWhat it does
Approved modelsPolicy > Model and endpoint accessDefault deny: a call is allowed only when the model is on the approved list. The list is platform AND tenant AND agent, and each layer can only narrow the one above.
Access rulesSame pageAllow or deny by provider, route type, model, hosted tools, function tools and output type, for the tenant, a department, a person or one agent. The narrowest rule wins, and a 7-day impact preview shows what a draft would change.
Sensitive dataDLPPer entity type: redact and record, record only, or block the request.
Budgets and capsBudgetsBudgets for the company, org, department, person and agent. Each request is priced and held against the budget before it leaves, and an emergency cap stops a runaway agent now.
ModePer controlOff, watch or enforce. Watch records what would have been refused and changes nothing on the wire.

Start in watch, read what would have been refused, then switch on enforce. See Watch Before You Enforce.

04 / INTEGRATIONS: AI MODEL PROVIDERS

What the record shows

Every Cloudflare AI Gateway call leaves one signed record, allowed or refused.

  • One row per call: the agent and the person who owns it, the tenant, the model, the decision and the policy that made it, latency, and the cost, token and cache figures.
  • Tamper-evident: each row is hash-chained to the one before and signed with an ed25519 key held in Vault Transit, so an edited or missing row shows.
  • Real cost: cost comes from the usage block in Cloudflare AI Gateway's own reply, priced from your price book in USD per million tokens.
  • No unpriced spend: an agent under a budget that calls a model with no price is refused, so a new model cannot slip past a cap.

Conversations can be opened and replayed from the record. See Session Replay and One-Click Audit Evidence.

05 / INTEGRATIONS: AI MODEL PROVIDERS

Related capabilities and goals

More AI model providers

Every provider below is governed through Argent.

  • OpenAI
  • Anthropic
  • Google Gemini
  • Azure OpenAI and Foundry
  • Amazon Bedrock
  • Mistral
  • DeepSeek
  • OpenRouter
  • GitHub Copilot
  • Kiro
  • Ollama
  • vLLM
  • NVIDIA NIM
  • LM Studio
  • Hugging Face TGI