Portkey vs Vigil
What Portkey does and what Vigil does, side by side. Every Portkey fact below was read on Portkey's own pages on 2026-10-09 and links to the page it came from; a feature its pages do not mention is marked “not stated”, never “no”. Where Portkey is stronger, this page says so.
What Portkey is, and what Vigil is
Portkey, in its own words: “AI Gateway, Observability, Guardrails, Governance, and Prompt Management, all in one platform.”portkey.ai. Gateway: the Portkey SDK, or the OpenAI SDK pointed at Portkey’s gateway URL with an x-portkey-provider header.
Vigil is a proxy for AI API calls. Change one base URL, add one header, and every call is logged with its cost, tokens, latency, errors and agent, then forwarded unchanged. Where the provider supports explicit prompt caching, Vigil can place the cache markers itself when optimisation is switched on. It stores no prompt text and caches no responses.
- Portkey says: Developer plan: "Not suitable for production workloads."portkey.ai
- Portkey says: Production plan: "Not recommended for organizations requiring custom security controls or data residency guarantees."portkey.ai
Feature by feature
The Portkey column was read on Portkey's own pages on 2026-10-09; each cell links to the page. “Not stated” means the pages read do not mention it, which is not the same as “no”. The Vigil column is read from Vigil's code and docs.
| Feature | Portkey | Vigil |
|---|---|---|
| Integration | "Integrate Portkey in just 3 lines of code"; OpenAI SDK with base_url set to the gateway.portkey.ai | Proxy. Change the base URL and add the X-Vigil-Key header; no SDK. |
| Open source | The gateway is MIT; the README lists key management, semantic caching, access control and PII redaction as enterprise-only.github.com | No. Vigil is not open source. |
| Hosting | Cloud, self-hosted open-source gateway, and private cloud deployments on AWS, Azure, GCP.portkey.ai | Cloud only: a Cloudflare Worker in front of the provider, a dashboard on Vercel, data in Supabase (Frankfurt, EU). |
| Cost per call | Yes: each log shows "tokens generated, thinking tokens and cost"; analytics across "21+ key metrics".portkey.ai | Yes: cost, tokens, latency, status and agent on every call, from a rate registry where every rate carries its source, date and provenance; a dash where no rate is published. |
| Providers | "Portkey lets you access 1,600+ LLMs via a unified API"; the README also says "45+ providers" and "250+ LLMs".portkey.ai | 17 providers routed by segment. |
| Prompt caching | Passes your markers through: "Just set the cache_control param in your respective message body"; no automatic injection described.portkey.ai | Creates cache hits: Anthropic, Google Vertex, Mistral get cache markers placed by Vigil when optimisation is on; AWS Bedrock when the request carries an API key rather than a SigV4 signature. On other providers Vigil records the provider’s own caching. |
| Response caching | Yes: simple (exact) caching on all plans; semantic caching "only available on select Enterprise plans".portkey.ai | No. Vigil does not cache responses. |
| Routing and fallbacks | Fallbacks, weighted load balancing across keys and providers, automatic retries up to 5 times.github.com | Model routing is in beta: Sends a conversation to a cheaper model from the same provider — one tier down — when its opening's size and tools are in a band the agent's routing level covers and the cheaper model can take the request; every call of the conversation gets the same answer. No provider fallbacks or load balancing. |
| Prompt management | Yes: versioning, labels, a playground and a Prompt API.portkey.ai | No prompt management or versioning. |
| Evals | Feedback values and weights on requests. No evaluation feature on the pages read.portkey.ai | No evaluations. A score can be sent for a call through the feedback endpoint. |
| Tracing | "Portkey supports request tracing".portkey.ai | One record per call. No spans or traces. |
| Alerts | Listed under the Production plan’s observability; the alerts doc returned 404 on the day read.portkey.ai | Cost spikes, failures and findings are flagged in the dashboard. Nothing is sent by email, Slack or push. |
| Prompt injection and PII | Guardrails scan for prompt injection and can redact PII, with partner checks; a denied request is "killed with a 446 status code". Available on all plans with more checks higher up.portkey.ai | Flags likely prompt injections and personal data in prompts, on every plan. Nothing is blocked or redacted. |
| Budgets and rate limits | Budgets in USD or tokens and rate limits per minute, hour or day: "Available on Enterprise plan and select Pro customers".portkey.ai | No spend caps or rate limits on your provider traffic. Optimisation can be switched off per agent. |
| Seats | RBAC and service-account keys on Production; SSO and organisation management on Enterprise. Seat counts not stated.portkey.ai | Seats per plan: 1 on Free, 3 on Plus, unlimited from Pro. Teammates are invited; there are no role tiers. |
| Compliance | "compliant with leading security standards, including SOC2, ISO27001, GDPR, and HIPAA".portkey.ai | No SOC 2 or ISO 27001 report. A DPA is published. |
| Data residency | Not stated on the pages read. | Account and telemetry stored in the EU (Supabase, Frankfurt). The proxy runs on Cloudflare’s edge. No prompt text or responses are stored. |
Where Portkey is stronger
- An open-source gateway you can self-host, and private cloud deployments.
- Response caching, fallbacks, weighted load balancing and retries.
- Prompt management with versioning and a playground.
- Guardrails that redact or block, budgets and rate limits, and SOC2, ISO27001 and HIPAA claims.
Where Vigil differs
- It creates prompt-cache hits rather than reporting them, on the providers whose caching lets a proxy place markers, and measures what caching would have saved on the others.
- Every rate it prices with carries its source, its retrieval date and a provenance label, and a cost it cannot stand behind is a dash, not a number.
- Monitoring is not gated by plan: every agent, every provider, on Free. Paid plans switch optimisation on.
- It does less: no response cache, no evals, no prompt management, no traces, no alerts outside the dashboard.
Pricing, side by side
Portkeyportkey.ai
- Open Source Self-host: no request limit; gateway, routing, guardrails, basic dashboard
- Developer Free: 10k recorded logs a month, 3 days of logs, 30 days of metrics, 3 prompt templates
- Production $49/mo: 100k recorded logs a month, $9 per additional 100k; 30 days of logs, 90 days of metrics; alerts, RBAC
- Enterprise Custom: 10 million+ logs, custom retention, SOC2 Type 2, private cloud, semantic caching
The page mixes "logs" and "requests" for the same limit; the figures are as printed.
Vigil vigil.tools/pricing
- Free $0: every agent monitored; 10,000 logged calls a month, 7 days of history; every optimisation On during your 7-day trial, as Plus, then Shadow only
- Plus $29/mo: every optimisation On up to $100 saved a billing month, then Shadow until the next; 500,000 logged calls, 30 days
- Pro $99/mo: every optimisation On up to $500 saved a billing month, then Shadow until the next; 2,000,000 logged calls, 90 days
- Supreme $299/mo: every optimisation On up to $2,000 saved a billing month, then Shadow until the next; Unlimited logged calls, 365 days
- Enterprise Custom: by contract
Monthly, excl. VAT. Monitoring is free on every plan; what a paid plan buys is a higher limit on the money Vigil saves you per billing month before optimisations go to Shadow.
Other comparisons: Helicone vs Vigil · LiteLLM vs Vigil · Langfuse vs Vigil · OpenRouter vs Vigil