Alternative

An OpenRouter alternative, compared honestly

Vigil is not a model router and does not replace OpenRouter’s one key for hundreds of models: it forwards to OpenRouter the way it forwards to any provider. This page is for the reader who used OpenRouter mainly for its per-request cost view and wants that, per agent, across providers they call directly. If you are looking at alternatives to OpenRouter, this page puts Vigil beside it without flattering either. Every OpenRouter fact was read on OpenRouter's own pages on 2026-10-09, its pricing and status again on 2026-11-16, and links to its source. Vigil does less than OpenRouter in several places, listed plainly below; what it does differently is a proxy that creates prompt-cache hits and prices every call with its provenance shown.

What OpenRouter is, and what Vigil is

OpenRouter, in its own words: “A unified API for every major LLM. One endpoint, hundreds of models.”openrouter.ai. Gateway: the OpenAI SDK with base_url https://openrouter.ai/api/v1, "a drop-in replacement".

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 unless you switch on an optimisation. Prompt caching, where the provider supports it, never changes answers. History trimming and model routing can change answers: both are opt-in, run in Shadow first, and stop automatically if errors, cut-off answers or failed answers rise. It stores no prompt text and caches no responses.

  • OpenRouter says: "We pass through the pricing of the underlying model providers without any markup."openrouter.ai
  • OpenRouter says: it does not store prompts or responses by default, and stores metadata such as token counts and latency.openrouter.ai

Feature by feature

The OpenRouter column was read on OpenRouter's own pages on 2026-10-09, and the cells marked “re-read” again on 2026-11-16; 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.

FeatureOpenRouterVigil
IntegrationOpenAI SDK pointed at openrouter.ai/api/v1.openrouter.aiProxy. Change the base URL and add the X-Vigil-Key header; no SDK.
Open sourceThe service is closed; SDKs are Apache-2.0.github.comNo. Vigil is not open source.
HostingCloud only, with eu.openrouter.ai and us.openrouter.ai regional hosts on Business and Enterprise.openrouter.aiCloud only: a Cloudflare Worker in front of the provider, a dashboard on Vercel, data in Supabase (Frankfurt, EU).
Cost per call"cost tracking across providers"; a cache_discount field and a generation endpoint per request.openrouter.aiYes: 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"Unified access to 500+ AI models" and "50+ cloud providers"; the public models API listed 458 models on the day read.openrouter.ai17 providers routed by segment.
Prompt caching"Most providers automatically enable prompt caching"; on Anthropic, a top-level cache_control option for automatic breakpoints or per-block markers you place.openrouter.aiCreates 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 cachingNot stated on the pages read.No. Vigil does not cache responses.
Routing and fallbacks"load balance requests across providers, prioritizing price"; fallbacks on by default; an Auto Router picks a model for the prompt.openrouter.aiModel routing is in beta: Sends a call to a cheaper model from the same provider — one tier down — when its answer is closed-form and checkable (a label, JSON or other data, a number, code to a named function or test, or a request with tools), its prompt is at most about 16,000 tokens, its opening's size and tools are in a band the agent's routing level covers and the cheaper model can take the request. Open-ended work without tools — writing, summaries, reviews, explanations, chat — stays on the original model, and once a call of a conversation stays there, the later ones do too; a later call the cheaper model can't take stays on the original model. Routing can change answers, so it runs in Shadow first. No provider fallbacks or load balancing.
Prompt managementNot stated on the pages read.No prompt management or versioning.
EvalsNot stated on the pages read.No evaluations. A score can be sent for a call through the feedback endpoint.
TracingBroadcast sends traces to 21 external platforms including Langfuse, LangSmith and Datadog.openrouter.aiOne record per call. No spans or traces.
AlertsBudget, key-spend and new-model alerts by email, Slack and webhook.openrouter.aiCost spikes, failures and findings are flagged in the dashboard. Nothing is sent by email, Slack or push.
Prompt injection and PIIGuardrails: regex-based prompt-injection detection; "Detect and redact or block sensitive information (PII)".openrouter.aiFlags likely prompt injections and personal data in prompts, on every plan. Nothing is blocked or redacted.
Budgets and rate limitsSpending caps in USD per day, week or month; model allowlists; per-key credit limits.openrouter.aiNo spend caps or rate limits on your provider traffic. Optimisation can be switched off per agent.
SeatsWorkspaces: 5, 5, 1,000 or custom by plan; SSO and SAML on Enterprise.openrouter.aiSeats per plan: 1 on Free, 3 on Plus, unlimited from Pro. Teammates are invited; there are no role tiers.
Compliance"A single GDPR-compatible, SOC 2 compliant partner with SLAs."openrouter.aiNo SOC 2 or ISO 27001 report. A DPA is published.
Data residencyIn-region routing on Business and Enterprise; a zero-data-retention policy unless you opt in to prompt logging.openrouter.aiAccount and telemetry stored in the EU (Supabase, Frankfurt). The proxy runs on Cloudflare’s edge. No prompt text or responses are stored.

Where OpenRouter is stronger

  • One key for hundreds of models across many providers, with load balancing and fallbacks.
  • Provider prices without markup, a free tier of models, and bring-your-own-key.
  • Spending caps, model allowlists and guardrails that redact or block.
  • EU or US in-region routing and a zero-data-retention default.

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

OpenRouteropenrouter.ai read 2026-11-16

  • Free No fee: 25+ free models, 50 requests a day
  • Standard Pay as you go: provider price without markup; 5.5% fee ($0.80 minimum) when buying credits; bring your own key free up to $25,000 of inference a month
  • Business Pay as you go: 8% platform fee; EU or US in-region routing; 1,000 workspaces
  • Enterprise Contact: fee discounts, SSO and SAML, contractual SLAs

Tiers re-read 2026-11-16: unchanged.

Vigil vigil.tools/pricing

  • Free $0: every agent monitored; 10,000 logged calls a month, 7 days of history; every optimisation On until $30 saved, once ever, then a 30-day grace, then Shadow only
  • Plus $29/mo: every optimisation On up to $100 saved a billing month, then a 30-day grace, then Shadow; 500,000 logged calls, 30 days
  • Pro $99/mo: every optimisation On up to $500 saved a billing month, then a 30-day grace, then Shadow; 2,000,000 logged calls, 90 days
  • Supreme $299/mo: every optimisation On up to $2,000 saved a billing month, then a 30-day grace, then Shadow; 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 period before optimisations go to Shadow (after a 30-day grace).

Moving from OpenRouter: what changes and what you keep

Vigil is a base-URL change and one header, so there is nothing to uninstall and nothing to instrument: the SDK you already use keeps its code. Your provider keys stay where they are. If you depend on OpenRouter for a response cache, routing, prompt management or traces, Vigil does not replace those; the table above says which. Vigil is not open source and runs only as a hosted service, so a self-hosting requirement rules it out.

What you gain is a proxy that can place prompt-cache markers for you on the providers that allow it, a cost on every call with its provenance shown, and monitoring that is free on every plan. Start on Free and compare your own numbers before switching anything off.

Questions

Does Vigil replace OpenRouter?
No. OpenRouter is one key for hundreds of models with load balancing and fallbacks; Vigil is a cost proxy that forwards to OpenRouter the way it forwards to any provider (see /integrations/openrouter). Use both when you want OpenRouter’s routing and a per-agent cost view across every provider you call, OpenRouter included.
Does OpenRouter charge a markup?
Not on inference: "We pass through the pricing of the underlying model providers without any markup." The fee is on credits: 5.5% ($0.80 minimum) when buying them on the Standard plan, 8% on Business, and bring-your-own-key is free up to $25,000 of inference a month (openrouter.ai/docs/faq and /pricing, read 2026-10-09, tiers re-read 2026-11-16).
Does OpenRouter show cost per agent?
Per request and per key: a cost on each generation, "cost tracking across providers", and spending caps per key (OpenRouter’s docs, read 2026-10-09). An agent is not a concept there. Vigil names each agent from a request header and prices its calls separately, across providers, which is the one thing this page is for.

Also: Helicone alternative · Portkey alternative · LiteLLM alternative · Langfuse alternative