{
  "name": "The Token Ledger",
  "generated": "2026-10-20",
  "license_note": "Facts compiled from public vendor sources; see source URLs.",
  "scoring": "score_pct = points / applicable max; yes=2, partial=1, no/unknown=0; n/a excluded. Criteria: Hard budgets (team / key / user), Rate & spend limits, Cost attribution / chargeback tags, FOCUS export, EU data residency option, Public pricing",
  "tools": [
    {
      "name": "LiteLLM",
      "slug": "litellm",
      "vendor": "BerriAI",
      "url": "https://www.litellm.ai",
      "category": "Gateway",
      "summary": "Open-source, self-hosted LLM proxy with virtual keys, multi-level budgets and an experimental FOCUS export.",
      "open_source": {
        "value": "yes",
        "detail": "MIT for the core; the enterprise/ directory is under a commercial license",
        "source": "https://github.com/BerriAI/litellm/blob/main/LICENSE"
      },
      "hosting": {
        "value": "Self-host",
        "detail": "Self-hosted or air-gapped in your own infrastructure",
        "source": "https://www.litellm.ai/enterprise"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Budgets per virtual key, team, team member, internal user, customer and agent; multiple budget windows; optional fail-closed enforcement. Requires a Postgres database; per-model budgets need Enterprise",
        "source": "https://docs.litellm.ai/docs/proxy/users"
      },
      "rate_limits": {
        "value": "yes",
        "detail": "RPM/TPM and max parallel requests per key, user, team, model and agent",
        "source": "https://docs.litellm.ai/docs/proxy/users"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Spend by key, user, team and org; tag-based spend tracking",
        "source": "https://www.litellm.ai/enterprise"
      },
      "focus_export": {
        "value": "yes",
        "detail": "Experimental export conforming to FOCUS 1.2 (Parquet/CSV to S3, GCS, Vantage; also Ternary and Mavvrik connectors)",
        "source": "https://docs.litellm.ai/docs/observability/focus"
      },
      "focus_version": "1.2",
      "providers": {
        "value": "100+",
        "detail": "\"Call 100+ LLM APIs\"",
        "source": "https://github.com/BerriAI/litellm"
      },
      "caching": {
        "value": "yes",
        "detail": "Redis/Valkey exact cache and Redis semantic cache",
        "source": "https://docs.litellm.ai/docs/proxy/caching_semantic"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "Self-hosted: runs in any region you choose",
        "source": "https://www.litellm.ai/enterprise"
      },
      "pricing_model": {
        "value": "Open source + Enterprise license",
        "source": "https://www.litellm.ai/enterprise"
      },
      "price_points": {
        "value": "OSS free; Enterprise: custom pricing (talk to sales), 30-day trial key",
        "source": "https://www.litellm.ai/enterprise"
      },
      "calc": {
        "type": "free",
        "note": "Open-source software: no license fee. Your own hosting, Postgres and Redis costs are not included."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://www.litellm.ai/enterprise"
      },
      "score_points": 12,
      "score_max": 12,
      "score_pct": 100,
      "providers_num": 100,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Portkey (Prisma AIRS AI Gateway)",
      "slug": "portkey",
      "vendor": "Palo Alto Networks (acquired Portkey)",
      "url": "https://portkey.ai",
      "category": "Gateway",
      "summary": "AI gateway with budget and usage-limit policies, now sold by Palo Alto Networks as Prisma AIRS AI Gateway.",
      "open_source": {
        "value": "partial",
        "detail": "Core gateway is MIT-licensed; governance, budgets and log export are in the commercial product",
        "source": "https://github.com/Portkey-AI/gateway"
      },
      "hosting": {
        "value": "Both",
        "detail": "SaaS or hybrid (data plane in your Kubernetes); OSS gateway can be self-hosted",
        "source": "https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/deploy-ai-gateway"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Usage-limit policies on cost, tokens or requests, grouped by metadata (e.g. per user); requests blocked with HTTP 412. Listed as Enterprise on the pricing page",
        "source": "https://portkey.ai/docs/product/enterprise-offering/budget-policies"
      },
      "rate_limits": {
        "value": "yes",
        "detail": "\"Granular Budget & Rate Limits\" in Enterprise",
        "source": "https://portkey.ai/pricing"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Metadata keys attribute cost to teams, users, features",
        "source": "https://portkey.ai/docs/product/observability/metadata"
      },
      "focus_export": {
        "value": "no",
        "detail": "No FOCUS export documented; JSONL log export (Enterprise only, 50k logs per job)",
        "source": "https://portkey.ai/docs/product/observability/logs-export"
      },
      "focus_version": null,
      "providers": {
        "value": "1,600+ LLMs (OSS) / 3,000+ LLMs (Prisma AIRS)",
        "detail": "Counts are models, not providers",
        "source": "https://www.paloaltonetworks.com/blog/ai-security/securing-and-governing-ai-agents-at-scale-through-a-unified-ai-gateway/"
      },
      "caching": {
        "value": "yes",
        "detail": "Simple and semantic caching (Production plan)",
        "source": "https://portkey.ai/pricing"
      },
      "eu_residency": {
        "value": "partial",
        "detail": "Portkey Enterprise docs describe EU region-pinned SaaS and hybrid; Prisma AIRS AI Gateway deploy docs currently name the Americas region",
        "source": "https://github.com/Portkey-AI/docs-core/blob/main/enterprise/security.mdx"
      },
      "pricing_model": {
        "value": "Freemium SaaS + Enterprise",
        "source": "https://portkey.ai/pricing"
      },
      "price_points": {
        "value": "Developer free (10k logs/mo); Production $49/mo incl. 100k logs, +$9 per extra 100k up to 3M; Enterprise custom",
        "source": "https://portkey.ai/pricing"
      },
      "calc": {
        "type": "portkey",
        "note": "Production plan: $49 incl. 100k recorded requests, +$9 per extra 100k, up to 3M. Above 3M only Enterprise (custom). Pricing page may change after the Palo Alto Networks acquisition."
      },
      "news": [
        {
          "value": "Palo Alto Networks closed its acquisition of Portkey on 29 May 2026; product is now Prisma AIRS AI Gateway (GA 16 Jul 2026).",
          "source": "https://www.paloaltonetworks.com/company/press/2026/palo-alto-networks-completes-acquisition-of-portkey-to-secure-ai-agents"
        }
      ],
      "pricing_public": {
        "value": "partial",
        "source": "https://portkey.ai/pricing"
      },
      "score_points": 8,
      "score_max": 12,
      "score_pct": 67,
      "providers_num": 1600,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Kong AI Gateway",
      "slug": "kong-ai-gateway",
      "vendor": "Kong Inc.",
      "url": "https://konghq.com/products/kong-ai-gateway",
      "category": "Gateway",
      "summary": "AI plugins on the Kong API gateway, sold via Konnect (usage-based) or Enterprise; cost-based limits per consumer group.",
      "open_source": {
        "value": "partial",
        "detail": "Kong Gateway is Apache-2.0 with free AI plugins (e.g. AI Proxy); AI Rate Limiting Advanced and AI Semantic Cache are paid plugins",
        "source": "https://konghq.com/blog/product-releases/announcing-kong-ai-gateway"
      },
      "hosting": {
        "value": "Both",
        "detail": "Konnect serverless, dedicated cloud (AWS/GCP/Azure region of choice) or hybrid; fully self-hosted via Gateway Enterprise",
        "source": "https://konghq.com/pricing"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "AI Rate Limiting Advanced (Enterprise plugin) enforces token or dollar-cost limits per consumer group",
        "source": "https://developer.konghq.com/plugins/ai-rate-limiting-advanced/"
      },
      "rate_limits": {
        "value": "yes",
        "detail": "Token and cost strategies; model input/output cost configured per 1M tokens",
        "source": "https://developer.konghq.com/ai-gateway/model-cost-management/"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Consumers and consumer groups; Konnect Metering & Billing lists showback and token budgeting",
        "source": "https://developer.konghq.com/ai-gateway/entities/ai-consumer-group/"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found in Kong docs",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "16",
        "detail": "AI Proxy list: Anthropic, Azure OpenAI, Bedrock, Cerebras, Cohere, DashScope, Databricks, DeepSeek, Gemini, Hugging Face, Llama, Mistral, Ollama, OpenAI, vLLM, xAI",
        "source": "https://developer.konghq.com/plugins/ai-proxy"
      },
      "caching": {
        "value": "yes",
        "detail": "AI Semantic Cache (Enterprise plugin): exact and semantic, Redis/Valkey or pgvector",
        "source": "https://developer.konghq.com/plugins/ai-semantic-cache/"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "Dedicated cloud gateways in a region of your choice; self-hosted option",
        "source": "https://konghq.com/pricing"
      },
      "pricing_model": {
        "value": "Usage-based (Konnect Plus) or custom Enterprise",
        "source": "https://konghq.com/pricing"
      },
      "price_points": {
        "value": "Konnect Plus: serverless AI gateway control plane $25/mo; hybrid $200/mo; dedicated $500/mo + $0.15/GB; 1M requests incl., $200 per extra 1M (max 10M/mo); $100/mo per proxied model; AI paid plugins as add-on (price not public)",
        "source": "https://konghq.com/pricing"
      },
      "calc": {
        "type": "kong",
        "note": "Konnect Plus, serverless control plane ($25) + 1M requests included + $200 per extra 1M (Plus max 10M) + $100 per proxied model per month. The AI Rate Limiting Advanced add-on price is not public and is not included."
      },
      "news": [],
      "pricing_public": {
        "value": "partial",
        "source": "https://konghq.com/pricing"
      },
      "score_points": 9,
      "score_max": 12,
      "score_pct": 75,
      "providers_num": 16,
      "last_checked": "2026-10-20"
    },
    {
      "name": "TrueFoundry AI Gateway",
      "slug": "truefoundry",
      "vendor": "TrueFoundry (Ensemble Labs Inc.)",
      "url": "https://www.truefoundry.com/ai-gateway",
      "category": "Gateway",
      "summary": "LLM and MCP gateway with team budget rules, per-user/team rate limits and transparent per-seat pricing.",
      "open_source": {
        "value": "no",
        "detail": "Commercial product; no open-source gateway listed",
        "source": "https://www.truefoundry.com/pricing"
      },
      "hosting": {
        "value": "Both",
        "detail": "SaaS; VPC, on-prem and air-gapped on Enterprise",
        "source": "https://www.truefoundry.com/pricing"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Dollar limits per team, user, model, virtual account, agent or metadata; daily to lifetime periods. Developer: 1 rule, Pro: up to 10",
        "source": "https://www.truefoundry.com/docs/ai-gateway/budget-limiting-v2"
      },
      "rate_limits": {
        "value": "yes",
        "detail": "Tokens/requests per minute, per user or aggregate per team",
        "source": "https://www.truefoundry.com/docs/ai-gateway/ratelimiting-v2"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Cost by user, team, virtual account, model and metadata",
        "source": "https://www.truefoundry.com/docs/ai-gateway/cost-tracking"
      },
      "focus_export": {
        "value": "no",
        "detail": "CSV/API metrics export; no FOCUS format documented",
        "source": "https://www.truefoundry.com/docs/ai-gateway/analytics"
      },
      "focus_version": null,
      "providers": {
        "value": "1,000+ LLMs",
        "detail": "Count is models",
        "source": "https://www.truefoundry.com/docs/ai-gateway/intro-to-llm-gateway"
      },
      "caching": {
        "value": "yes",
        "detail": "Exact and semantic caching",
        "source": "https://www.truefoundry.com/docs/ai-gateway/caching"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "Data residency on Enterprise",
        "source": "https://www.truefoundry.com/pricing"
      },
      "pricing_model": {
        "value": "Per seat + request overage",
        "source": "https://www.truefoundry.com/pricing"
      },
      "price_points": {
        "value": "Developer $0 (≤3 users, 10k req/user/mo); Pro $25/user/mo incl. 20k req/user (pooled), $20 per extra 100k; Enterprise custom annual",
        "source": "https://www.truefoundry.com/pricing"
      },
      "calc": {
        "type": "truefoundry",
        "note": "Pro plan: $25 per user + $20 per started 100k requests above the pooled 20k/user allowance (matches the vendor's own calculator)."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://www.truefoundry.com/pricing"
      },
      "score_points": 10,
      "score_max": 12,
      "score_pct": 83,
      "providers_num": 1000,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Cloudflare AI Gateway",
      "slug": "cloudflare-ai-gateway",
      "vendor": "Cloudflare",
      "url": "https://developers.cloudflare.com/ai-gateway/",
      "category": "Gateway",
      "summary": "Edge-hosted AI gateway; core features free, with spend limits by provider, model or custom metadata.",
      "open_source": {
        "value": "no",
        "detail": "Managed Cloudflare service",
        "source": "https://developers.cloudflare.com/ai-gateway/reference/pricing/"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://developers.cloudflare.com/ai-gateway/reference/pricing/"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "USD spend limits scoped by provider, model and custom metadata (e.g. per user/key), fixed or sliding windows; block or fall back",
        "source": "https://developers.cloudflare.com/ai-gateway/features/spend-limits/"
      },
      "rate_limits": {
        "value": "yes",
        "detail": "Rate limiting is a free core feature",
        "source": "https://developers.cloudflare.com/ai-gateway/reference/pricing/"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Custom metadata per request",
        "source": "https://developers.cloudflare.com/ai-gateway/observability/custom-metadata/"
      },
      "focus_export": {
        "value": "no",
        "detail": "No FOCUS export documented; logs via Logpush (billed)",
        "source": "https://developers.cloudflare.com/ai-gateway/reference/pricing/"
      },
      "focus_version": null,
      "providers": {
        "value": "23",
        "detail": "Provider-native endpoints listed in docs",
        "source": "https://developers.cloudflare.com/ai-gateway/usage/providers/"
      },
      "caching": {
        "value": "yes",
        "detail": "Exact-match caching of identical requests",
        "source": "https://developers.cloudflare.com/ai-gateway/features/caching/"
      },
      "eu_residency": {
        "value": "unknown",
        "detail": "No EU-only processing option documented; a community thread asks for it: https://community.cloudflare.com/t/ai-gateway-any-roadmap-for-data-residency-regional-processing-regional-services/933015",
        "source": null
      },
      "pricing_model": {
        "value": "Free core + paid logs / billing fee",
        "source": "https://developers.cloudflare.com/ai-gateway/reference/pricing/"
      },
      "price_points": {
        "value": "Core features free on all plans; logs follow Workers Logs pricing (accounts created after 24 Sep 2026); Unified Billing adds 5% on credits; no markup on inference",
        "source": "https://developers.cloudflare.com/ai-gateway/reference/pricing/"
      },
      "calc": {
        "type": "cloudflare",
        "note": "Core gateway is free. If you pay providers through Unified Billing, a 5% fee on credits applies (toggle). Log storage/Logpush costs not included."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://developers.cloudflare.com/ai-gateway/reference/pricing/"
      },
      "score_points": 8,
      "score_max": 12,
      "score_pct": 67,
      "providers_num": 23,
      "last_checked": "2026-10-20"
    },
    {
      "name": "OpenRouter",
      "slug": "openrouter",
      "vendor": "OpenRouter",
      "url": "https://openrouter.ai",
      "category": "Gateway",
      "summary": "Hosted model marketplace and router: one API, provider list prices, a fee on credit purchases.",
      "open_source": {
        "value": "no",
        "detail": "Hosted service",
        "source": "https://openrouter.ai/pricing"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://openrouter.ai/pricing"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Budgets & spend controls (Standard and up); each API key can carry its own credit cap; workspaces per environment",
        "source": "https://openrouter.ai/pricing"
      },
      "rate_limits": {
        "value": "partial",
        "detail": "No plan-based limits on paid models (provider limits pass through); per-key credit caps act as spend limits",
        "source": "https://openrouter.ai/pricing"
      },
      "cost_attribution": {
        "value": "partial",
        "detail": "Per-key and per-workspace activity, logs and export",
        "source": "https://openrouter.ai/pricing"
      },
      "focus_export": {
        "value": "no",
        "detail": "Activity logs & export; no FOCUS format documented",
        "source": "https://openrouter.ai/pricing"
      },
      "focus_version": null,
      "providers": {
        "value": "80+ providers (500+ models)",
        "source": "https://openrouter.ai/pricing"
      },
      "caching": {
        "value": "partial",
        "detail": "Provider prompt caching (Standard and up); no gateway response cache listed",
        "source": "https://openrouter.ai/pricing"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "EU in-region routing via eu.openrouter.ai on Business and Enterprise",
        "source": "https://openrouter.ai/pricing"
      },
      "pricing_model": {
        "value": "Fee on credit purchases",
        "source": "https://openrouter.ai/pricing"
      },
      "price_points": {
        "value": "Standard 5.5% fee on credits; Business 8%; Enterprise discounts; inference at provider list price; BYOK: first $25k/mo list-price inference free, then 5%",
        "source": "https://openrouter.ai/pricing"
      },
      "calc": {
        "type": "openrouter",
        "note": "Standard plan: 5.5% fee on credits bought (≈ your token spend). BYOK mode: 0% up to $25k/month list-price inference, then 5%."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://openrouter.ai/pricing"
      },
      "score_points": 8,
      "score_max": 12,
      "score_pct": 67,
      "providers_num": 80,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Requesty",
      "slug": "requesty",
      "vendor": "Requesty",
      "url": "https://www.requesty.ai",
      "category": "Gateway",
      "summary": "Hosted LLM gateway for 600+ models with monthly spend caps per key and user, an EU endpoint in Frankfurt and a 5% fee on model cost (0% with your own keys).",
      "open_source": {
        "value": "no",
        "detail": "Managed platform; no open-source gateway found",
        "source": "https://www.requesty.ai/enterprise"
      },
      "hosting": {
        "value": "SaaS",
        "detail": "Enterprise FAQ: \"Can Requesty be self-hosted? Not today.\" Regional endpoints: Global, EU (Frankfurt), US, AP",
        "source": "https://www.requesty.ai/enterprise"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Monthly spend caps per API key, service account and user; \"new requests for that key are blocked\" at the limit. Group/team budgets (shared or per member) are Enterprise only (https://docs.requesty.ai/features/groups)",
        "source": "https://docs.requesty.ai/features/cost-tracking"
      },
      "rate_limits": {
        "value": "partial",
        "detail": "No per-minute limits: one org-wide in-flight concurrency limit set by tier and balance (HTTP 429); per-key/user monthly spend caps act as spend limits",
        "source": "https://docs.requesty.ai/features/in-flight-rate-limits"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Cost by model, provider, user, member, API key and custom request-metadata fields; per-request cost in the API response; CSV export",
        "source": "https://docs.requesty.ai/features/cost-tracking"
      },
      "focus_export": {
        "value": "no",
        "detail": "CSV export (pivot format) from the Advanced analytics tab; no FOCUS format documented",
        "source": "https://docs.requesty.ai/features/cost-tracking"
      },
      "focus_version": null,
      "providers": {
        "value": "20+ providers (600+ models)",
        "detail": "Startup program page says 30+ providers (https://www.requesty.ai/startup)",
        "source": "https://www.requesty.ai/pricing"
      },
      "caching": {
        "value": "partial",
        "detail": "Auto caching adds provider prompt-cache breakpoints; docs say it is not server-side response storage",
        "source": "https://docs.requesty.ai/features/auto-caching"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "EU endpoint router.eu.requesty.ai: processing and storage in Frankfurt (AWS eu-central-1); org-wide EU-only enforcement; inference stays in the EU only with EU-region models",
        "source": "https://docs.requesty.ai/features/eu-routing"
      },
      "pricing_model": {
        "value": "Markup on model cost (pay as you go) + Enterprise",
        "source": "https://www.requesty.ai/pricing"
      },
      "price_points": {
        "value": "Free $0 (free models, 200 requests/day); Pay as you go: 5% on model cost (a 5% service fee on credit top-ups), no seat fee or minimum; BYOK: 0%; Enterprise custom",
        "detail": "BYOK 0% is stated on https://www.requesty.ai/enterprise; top-up fee on https://docs.requesty.ai/features/billing",
        "source": "https://www.requesty.ai/pricing"
      },
      "calc": {
        "type": "requesty",
        "note": "Pay as you go: 5% on model cost (≈ your token spend). BYOK mode: 0% gateway fee (provider bills you directly). Enterprise terms not public."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://www.requesty.ai/pricing"
      },
      "score_points": 9,
      "score_max": 12,
      "score_pct": 75,
      "providers_num": 20,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Azure API Management (AI gateway)",
      "slug": "azure-apim-ai-gateway",
      "vendor": "Microsoft",
      "url": "https://learn.microsoft.com/en-us/azure/api-management/genai-gateway-capabilities",
      "category": "Gateway",
      "summary": "AI gateway policies in Azure API Management: token limits and quotas, token metrics with custom dimensions, semantic caching.",
      "open_source": {
        "value": "no",
        "detail": "Managed Azure service",
        "source": "https://azure.microsoft.com/en-us/pricing/details/api-management/"
      },
      "hosting": {
        "value": "Both",
        "detail": "Managed tiers; self-hosted gateway on Premium (and Developer for evaluation)",
        "source": "https://azure.microsoft.com/en-us/pricing/details/api-management/"
      },
      "hard_budgets": {
        "value": "partial",
        "detail": "Token quotas per counter key (e.g. subscription) per period, HTTP 403 when exceeded; quotas are in tokens, not currency",
        "source": "https://learn.microsoft.com/en-us/azure/api-management/llm-token-limit-policy"
      },
      "rate_limits": {
        "value": "yes",
        "detail": "Tokens per minute per counter key (HTTP 429)",
        "source": "https://learn.microsoft.com/en-us/azure/api-management/llm-token-limit-policy"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Token metrics to Application Insights with up to 5 custom dimensions",
        "source": "https://learn.microsoft.com/en-us/azure/api-management/llm-emit-token-metric-policy"
      },
      "focus_export": {
        "value": "partial",
        "detail": "Azure Cost Management exports the Azure bill in FOCUS; per-consumer token data from APIM is not in that export",
        "source": "https://focus.finops.org/docs/implementation/data-generators/microsoft-azure/"
      },
      "focus_version": "1.2 (Azure bill)",
      "providers": {
        "value": "unknown",
        "detail": "Azure OpenAI / Foundry models and OpenAI-compatible APIs; no count published",
        "source": null
      },
      "caching": {
        "value": "yes",
        "detail": "Semantic caching policies",
        "source": "https://learn.microsoft.com/en-us/azure/api-management/genai-gateway-capabilities"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "Deployable in EU regions (e.g. West Europe)",
        "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'API%20Management'%20and%20armRegionName%20eq%20'westeurope'"
      },
      "pricing_model": {
        "value": "Per unit-hour + per-request overage",
        "source": "https://azure.microsoft.com/en-us/pricing/details/api-management/"
      },
      "price_points": {
        "value": "West Europe retail: Consumption first 1M calls free then $3.50/1M; Basic v2 ≈$150/mo (10M incl., $3/1M extra); Standard v2 ≈$700/mo (50M incl., $2.50/1M); new 'AI Gateway tier': price 'coming soon'",
        "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'API%20Management'%20and%20armRegionName%20eq%20'westeurope'"
      },
      "calc": {
        "type": "azure",
        "note": "Basic v2, West Europe retail: $0.20548/hour × 730 h ≈ $150/month, 10M requests included, $3.00 per extra 1M. Standard v2 or Consumption can be cheaper/dearer depending on volume. The new AI Gateway tier has no published price yet."
      },
      "news": [
        {
          "value": "Azure pricing page lists a new 'API Management (AI Gateway tier)' with 'Pricing details are coming soon'.",
          "source": "https://azure.microsoft.com/en-us/pricing/details/api-management/"
        }
      ],
      "pricing_public": {
        "value": "yes",
        "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'API%20Management'%20and%20armRegionName%20eq%20'westeurope'"
      },
      "score_points": 10,
      "score_max": 12,
      "score_pct": 83,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Google Cloud Apigee (AI gateway) + Vertex AI controls",
      "slug": "google-apigee-vertex",
      "vendor": "Google Cloud",
      "url": "https://docs.cloud.google.com/apigee/docs/api-platform/fundamentals/build-ai-gateway",
      "category": "Gateway",
      "summary": "Apigee LLM token policies, semantic cache and Model Armor in front of models; Vertex/Gemini request labels flow into Cloud Billing.",
      "open_source": {
        "value": "no",
        "detail": "Managed Google Cloud service",
        "source": "https://cloud.google.com/apigee/pricing"
      },
      "hosting": {
        "value": "SaaS",
        "detail": "Managed Apigee (hybrid option not reviewed here)",
        "source": "https://cloud.google.com/apigee/pricing"
      },
      "hard_budgets": {
        "value": "partial",
        "detail": "LLMTokenQuota: token quotas per minute/hour/day/week/month (HTTP 429); tokens, not currency",
        "source": "https://docs.cloud.google.com/apigee/docs/api-platform/reference/policies/llm-token-quota-policy"
      },
      "rate_limits": {
        "value": "yes",
        "source": "https://docs.cloud.google.com/apigee/docs/api-platform/reference/policies/llm-token-quota-policy"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Custom labels on Gemini API calls are forwarded to Cloud Billing and BigQuery export",
        "source": "https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/capabilities/add-labels-to-api-calls"
      },
      "focus_export": {
        "value": "partial",
        "detail": "Cloud Billing FOCUS BigQuery export (preview) covers the Google Cloud bill; Apigee per-consumer token data is not in it",
        "source": "https://docs.cloud.google.com/billing/docs/how-to/export-data-bigquery-tables/focus-export"
      },
      "focus_version": "1.2 preview (GCP bill)",
      "providers": {
        "value": "unknown",
        "detail": "Apigee can front any HTTP model API; no count published",
        "source": null
      },
      "caching": {
        "value": "yes",
        "detail": "Semantic cache in the reference AI gateway",
        "source": "https://docs.cloud.google.com/apigee/docs/api-platform/fundamentals/build-ai-gateway"
      },
      "eu_residency": {
        "value": "unknown",
        "detail": "EU regions not verified for this page",
        "source": null
      },
      "pricing_model": {
        "value": "Pay-as-you-go per call + environment fee, or subscription",
        "source": "https://cloud.google.com/apigee/pricing"
      },
      "price_points": {
        "value": "Extensible proxy (needed for LLM token policies): $100/1M calls up to 50M, $80 50–500M, $64 above; Intermediate environment $1,460/mo/region; Standard proxy $20/1M",
        "source": "https://cloud.google.com/apigee/pricing"
      },
      "calc": {
        "type": "apigee",
        "note": "Pay-as-you-go: Intermediate environment $1,460/month (one region) + Extensible proxy calls, tiered $100/$80/$64 per 1M. LLMTokenQuota is an Extensible policy. Networking not included."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://cloud.google.com/apigee/pricing"
      },
      "score_points": 8,
      "score_max": 12,
      "score_pct": 67,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Amazon Bedrock (native cost controls)",
      "slug": "aws-bedrock-controls",
      "vendor": "Amazon Web Services",
      "url": "https://docs.aws.amazon.com/bedrock/latest/userguide/cost-mgmt-application-inference-profiles.html",
      "category": "Gateway",
      "summary": "Not a separate gateway: Bedrock's own controls (inference profiles with cost allocation tags, Projects, AWS Budgets).",
      "open_source": {
        "value": "no",
        "detail": "Managed AWS service",
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/cost-mgmt-application-inference-profiles.html"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/cost-mgmt-application-inference-profiles.html"
      },
      "hard_budgets": {
        "value": "partial",
        "detail": "AWS Budgets alert (daily data) but do not stop usage; hard stops need a Lambda that attaches an IAM deny",
        "source": "https://repost.aws/articles/AR5rBgrAyQStqFOiBYu88PDg/governing-amazon-bedrock-costs-team-and-user-level-tracking-budgets-and-hard-spend-controls-for-claude-code"
      },
      "rate_limits": {
        "value": "unknown",
        "detail": "Per-team limits not documented in the sources reviewed",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Tagged application inference profiles appear in Cost Explorer/CUR (up to 24h lag); Bedrock Projects for Responses/Chat Completions APIs",
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/cost-mgmt-application-inference-profiles.html"
      },
      "focus_export": {
        "value": "yes",
        "detail": "AWS Data Exports FOCUS 1.2 (GA) covers Bedrock spend on the AWS bill",
        "source": "https://aws.amazon.com/about-aws/whats-new/2025/11/aws-data-exports-focus-1-2-available/"
      },
      "focus_version": "1.2 (AWS bill)",
      "providers": {
        "value": "unknown",
        "detail": "Multiple model providers; count not verified",
        "source": null
      },
      "caching": {
        "value": "yes",
        "detail": "Prompt caching (model-side)",
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-caching.html"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "EU geographic cross-region inference keeps processing within EU regions",
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/geographic-cross-region-inference.html"
      },
      "pricing_model": {
        "value": "No separate gateway product; pay per model usage",
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/cost-mgmt-application-inference-profiles.html"
      },
      "price_points": {
        "value": "n/a (controls are part of Bedrock/AWS Billing)",
        "source": null
      },
      "calc": {
        "type": "none",
        "note": "No gateway fee to compute: these are native Bedrock/AWS controls."
      },
      "news": [],
      "pricing_public": {
        "value": "n/a",
        "source": null
      },
      "score_points": 7,
      "score_max": 10,
      "score_pct": 70,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Vercel AI Gateway",
      "slug": "vercel-ai-gateway",
      "vendor": "Vercel",
      "url": "https://vercel.com/ai-gateway",
      "category": "Gateway",
      "summary": "Hosted gateway with zero token markup, budgets per team/project/key/member and paid add-ons for reporting.",
      "open_source": {
        "value": "no",
        "detail": "Hosted Vercel service",
        "source": "https://vercel.com/docs/ai-gateway/pricing"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://vercel.com/docs/ai-gateway/pricing"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Budgets cap spend for a team, project, API key or team member",
        "source": "https://vercel.com/docs/ai-gateway/pricing"
      },
      "rate_limits": {
        "value": "unknown",
        "detail": "Platform rate limits documented; custom per-team rate limits not verified",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Custom Reporting: tags, user IDs, quota entity IDs (paid add-on)",
        "source": "https://vercel.com/docs/ai-gateway/pricing"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "70+ providers (407 models)",
        "source": "https://vercel.com/ai-gateway/models"
      },
      "caching": {
        "value": "partial",
        "detail": "Automatic provider prompt caching",
        "source": "https://vercel.com/docs/ai-gateway/models-and-providers/automatic-caching"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "inferenceRegion 'eu' (fails rather than falls back if unsupported)",
        "source": "https://vercel.com/docs/ai-gateway/security-and-compliance/regional-inference"
      },
      "pricing_model": {
        "value": "No markup; paid add-ons",
        "source": "https://vercel.com/docs/ai-gateway/pricing"
      },
      "price_points": {
        "value": "Tokens at provider list price, no platform fee (incl. BYOK); Custom Reporting $0.075/1k writes + $5/1k queries; team-wide provider allowlist $0.10/1k requests; team-wide ZDR $0.10/1k requests; trace drains $0.05/1k traces",
        "source": "https://vercel.com/docs/ai-gateway/pricing"
      },
      "calc": {
        "type": "vercel",
        "note": "No gateway fee on tokens. If you turn on Custom Reporting for chargeback, assumed 1 tag write per request at $0.075 per 1,000 (query fees not included)."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://vercel.com/docs/ai-gateway/pricing"
      },
      "score_points": 8,
      "score_max": 12,
      "score_pct": 67,
      "providers_num": 70,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Bifrost",
      "slug": "bifrost",
      "vendor": "Maxim AI",
      "url": "https://getbifrost.ai",
      "category": "Gateway",
      "summary": "Apache-2.0 Go gateway with virtual keys, hierarchical budgets and semantic caching in the open-source edition.",
      "open_source": {
        "value": "yes",
        "detail": "Apache-2.0",
        "source": "https://github.com/maximhq/bifrost"
      },
      "hosting": {
        "value": "Self-host",
        "detail": "OSS self-host; Enterprise adds private/VPC deployment",
        "source": "https://getbifrost.ai/pricing"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Virtual keys with budgets; hierarchical budgets in OSS",
        "source": "https://docs.getbifrost.ai/features/governance/virtual-keys"
      },
      "rate_limits": {
        "value": "yes",
        "source": "https://docs.getbifrost.ai/features/governance/virtual-keys"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Budgets/usage per virtual key, team and customer",
        "source": "https://getbifrost.ai/resources/governance"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "20+ (docs) / 34 in provider matrix",
        "source": "https://docs.getbifrost.ai/providers/supported-providers/overview"
      },
      "caching": {
        "value": "yes",
        "detail": "Semantic caching",
        "source": "https://github.com/maximhq/bifrost"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "Self-hosted: runs where you deploy it",
        "source": "https://github.com/maximhq/bifrost"
      },
      "pricing_model": {
        "value": "Open source + Enterprise",
        "source": "https://getbifrost.ai/pricing"
      },
      "price_points": {
        "value": "OSS free; Enterprise custom (14-day trial)",
        "source": "https://getbifrost.ai/pricing"
      },
      "calc": {
        "type": "free",
        "note": "Open-source: no license fee. Hosting costs not included."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://getbifrost.ai/pricing"
      },
      "score_points": 10,
      "score_max": 12,
      "score_pct": 83,
      "providers_num": 20,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Envoy AI Gateway",
      "slug": "envoy-ai-gateway",
      "vendor": "Envoy Proxy project (CNCF)",
      "url": "https://github.com/envoyproxy/ai-gateway",
      "category": "Gateway",
      "summary": "Open-source Kubernetes AI gateway on Envoy Gateway; token-cost based rate limits per tenant.",
      "open_source": {
        "value": "yes",
        "detail": "Apache-2.0",
        "source": "https://github.com/envoyproxy/ai-gateway"
      },
      "hosting": {
        "value": "Self-host",
        "detail": "Kubernetes",
        "source": "https://github.com/envoyproxy/ai-gateway/blob/main/README.md"
      },
      "hard_budgets": {
        "value": "partial",
        "detail": "Usage-based rate limits on token costs per user/tenant header act as token budgets; no currency budgets",
        "source": "https://github.com/envoyproxy/ai-gateway/blob/main/examples/token_ratelimit/token_ratelimit.yaml"
      },
      "rate_limits": {
        "value": "yes",
        "detail": "llmRequestCosts (input/output/total/cached/CEL) + global rate limit",
        "source": "https://github.com/envoyproxy/ai-gateway/blob/main/examples/token_ratelimit/token_ratelimit.yaml"
      },
      "cost_attribution": {
        "value": "partial",
        "detail": "Token usage keyed by headers/metadata; no chargeback reporting",
        "source": "https://github.com/envoyproxy/ai-gateway/blob/main/examples/token_ratelimit/token_ratelimit.yaml"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "unknown",
        "detail": "README names OpenAI, Azure OpenAI, Gemini, Vertex AI, Bedrock, Mistral and others; no count",
        "source": null
      },
      "caching": {
        "value": "unknown",
        "source": null
      },
      "eu_residency": {
        "value": "yes",
        "detail": "Self-hosted",
        "source": "https://github.com/envoyproxy/ai-gateway"
      },
      "pricing_model": {
        "value": "Open source",
        "source": "https://github.com/envoyproxy/ai-gateway"
      },
      "price_points": {
        "value": "Free",
        "source": "https://github.com/envoyproxy/ai-gateway"
      },
      "calc": {
        "type": "free",
        "note": "Open-source: no license fee. Kubernetes and Redis costs not included."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://github.com/envoyproxy/ai-gateway"
      },
      "score_points": 8,
      "score_max": 12,
      "score_pct": 67,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Solo Enterprise for agentgateway (ex-Gloo AI Gateway)",
      "slug": "solo-agentgateway",
      "vendor": "Solo.io",
      "url": "https://docs.solo.io/agentgateway/latest/",
      "category": "Gateway",
      "summary": "Kubernetes AI gateway with declarative USD/token budgets per key, user, group, model or cost center (Enterprise).",
      "open_source": {
        "value": "partial",
        "detail": "Gloo is Apache-2.0; budgets and Cost Management UI need a Solo Enterprise license",
        "source": "https://github.com/solo-io/gloo"
      },
      "hosting": {
        "value": "Self-host",
        "detail": "Kubernetes",
        "source": "https://docs.solo.io/agentgateway/latest/install/licensing/"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "EnterpriseAgentgatewayBudget: USD or tokens, rolling day/week/month/year, Block (429) or Audit; fails open if rate-limit service is down",
        "source": "https://docs.solo.io/agentgateway/latest/llm/cost-controls/budget-limits/"
      },
      "rate_limits": {
        "value": "yes",
        "source": "https://docs.solo.io/agentgateway/kubernetes/latest/documentation/llm/rate-limit/"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Budget dimensions: virtual key, user, group, model, provider, custom (e.g. cost center)",
        "source": "https://docs.solo.io/agentgateway/latest/llm/cost-controls/budget-limits/"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "unknown",
        "source": null
      },
      "caching": {
        "value": "unknown",
        "source": null
      },
      "eu_residency": {
        "value": "yes",
        "detail": "Self-hosted",
        "source": "https://docs.solo.io/agentgateway/latest/install/licensing/"
      },
      "pricing_model": {
        "value": "Enterprise license",
        "source": "https://docs.solo.io/agentgateway/latest/install/licensing/"
      },
      "price_points": {
        "value": "Not public",
        "source": "https://docs.solo.io/agentgateway/latest/install/licensing/"
      },
      "calc": {
        "type": "custom"
      },
      "news": [],
      "pricing_public": {
        "value": "no",
        "source": "https://docs.solo.io/agentgateway/latest/install/licensing/"
      },
      "score_points": 8,
      "score_max": 12,
      "score_pct": 67,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Kosmoy",
      "slug": "kosmoy",
      "vendor": "Kosmoy (Milan, Italy)",
      "url": "https://www.kosmoy.com",
      "category": "Gateway",
      "summary": "EU-based, self-hosted AI control plane: gateway, budgets that warn then block, attribution and EU AI Act tooling.",
      "open_source": {
        "value": "no",
        "source": "https://www.kosmoy.com/resources/product-facts/"
      },
      "hosting": {
        "value": "Self-host",
        "detail": "Single-tenant in your Kubernetes (Azure, AWS, GCP, on-prem, air-gapped); self-hosted only",
        "source": "https://www.kosmoy.com/resources/product-facts/"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Budgets warn and then block at the gateway",
        "source": "https://www.kosmoy.com/resources/product-facts/"
      },
      "rate_limits": {
        "value": "unknown",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Every routed call attributed to provider, model, application, team, user, agent",
        "source": "https://www.kosmoy.com/resources/product-facts/"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "CSV/API exports described; FOCUS not mentioned (https://www.kosmoy.com/platform/cost-tracking/)",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "8+ named",
        "detail": "OpenAI, Anthropic, Google, Mistral, Meta, Azure, AWS, on-prem models",
        "source": "https://www.kosmoy.com/resources/product-facts/"
      },
      "caching": {
        "value": "unknown",
        "source": null
      },
      "eu_residency": {
        "value": "yes",
        "detail": "Self-hosted; prompts and logs never pass a Kosmoy-hosted service",
        "source": "https://www.kosmoy.com/resources/product-facts/"
      },
      "pricing_model": {
        "value": "Quote-based",
        "source": "https://www.kosmoy.com/resources/product-facts/"
      },
      "price_points": {
        "value": "No list price, no self-service plan",
        "source": "https://www.kosmoy.com/resources/product-facts/"
      },
      "calc": {
        "type": "custom"
      },
      "news": [],
      "pricing_public": {
        "value": "no",
        "source": "https://www.kosmoy.com/resources/product-facts/"
      },
      "score_points": 6,
      "score_max": 12,
      "score_pct": 50,
      "providers_num": 8,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Helicone",
      "slug": "helicone",
      "vendor": "Helicone",
      "url": "https://www.helicone.ai",
      "category": "Observability",
      "summary": "Open-source LLM observability with an AI gateway; cost-based rate limits per user or property. In maintenance mode.",
      "open_source": {
        "value": "yes",
        "detail": "Apache-2.0",
        "source": "https://github.com/Helicone/helicone"
      },
      "hosting": {
        "value": "Both",
        "detail": "Cloud; on-prem on Enterprise; OSS self-host",
        "source": "https://www.helicone.ai/pricing"
      },
      "hard_budgets": {
        "value": "partial",
        "detail": "Cost-based limits (in cents) per time window, per user or custom property",
        "source": "https://docs.helicone.ai/features/advanced-usage/custom-rate-limits"
      },
      "rate_limits": {
        "value": "yes",
        "source": "https://docs.helicone.ai/features/advanced-usage/custom-rate-limits"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Custom properties for cost by user/feature",
        "source": "https://docs.helicone.ai/features/advanced-usage/custom-properties"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "100+",
        "source": "https://docs.helicone.ai/gateway/overview"
      },
      "caching": {
        "value": "yes",
        "detail": "Response caching via headers",
        "source": "https://docs.helicone.ai/helicone-headers/header-directory"
      },
      "eu_residency": {
        "value": "yes",
        "detail": "EU region via eu- prefixed keys",
        "source": "https://docs.helicone.ai/helicone-headers/helicone-auth"
      },
      "pricing_model": {
        "value": "Freemium + usage-based",
        "source": "https://www.helicone.ai/pricing"
      },
      "price_points": {
        "value": "Free 10k requests; Pro $79/mo; Team $799/mo; usage-based pricing on top (rates not shown in page text)",
        "source": "https://www.helicone.ai/pricing"
      },
      "calc": {
        "type": "partial",
        "base": 79,
        "note": "Pro base fee $79/month; usage-based charges apply but rates are not published in text form."
      },
      "news": [
        {
          "value": "Acquired by Mintlify on 3 Mar 2026; product moved to maintenance mode (no new features).",
          "source": "https://www.helicone.ai/blog/joining-mintlify"
        },
        {
          "value": "As of Oct 2026 the Helicone site banner says 'Helicone is no longer part of Mintlify. Details to come soon.'",
          "source": "https://www.helicone.ai/about"
        }
      ],
      "pricing_public": {
        "value": "partial",
        "source": "https://www.helicone.ai/pricing"
      },
      "score_points": 8,
      "score_max": 12,
      "score_pct": 67,
      "providers_num": 100,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Langfuse",
      "slug": "langfuse",
      "vendor": "Langfuse (ClickHouse, Inc.)",
      "url": "https://langfuse.com",
      "category": "Observability",
      "summary": "Open-source LLM tracing and evals with token and cost tracking by user and tag; EU cloud region.",
      "open_source": {
        "value": "yes",
        "detail": "MIT, except ee/ directories",
        "source": "https://github.com/langfuse/langfuse/blob/main/LICENSE"
      },
      "hosting": {
        "value": "Both",
        "detail": "Langfuse Cloud or self-host",
        "source": "https://langfuse.com/pricing"
      },
      "hard_budgets": {
        "value": "n/a",
        "detail": "Observability only, not in the request path",
        "source": null
      },
      "rate_limits": {
        "value": "n/a",
        "detail": "Not in the request path",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Cost and tokens aggregated by user and tags; Metrics API",
        "source": "https://langfuse.com/docs/observability/features/token-and-cost-tracking"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "n/a",
        "detail": "SDK/OpenTelemetry instrumentation, not a proxy",
        "source": null
      },
      "caching": {
        "value": "n/a",
        "source": null
      },
      "eu_residency": {
        "value": "yes",
        "detail": "EU cloud region (AWS eu-west-1, Ireland)",
        "source": "https://langfuse.com/security/data-regions"
      },
      "pricing_model": {
        "value": "Open source + cloud tiers",
        "source": "https://langfuse.com/pricing"
      },
      "price_points": {
        "value": "Cloud Core $29/mo, Pro $199/mo, each incl. 100k units; overage from $8 per 100k units; self-host free",
        "source": "https://langfuse.com/pricing"
      },
      "calc": {
        "type": "langfuse",
        "note": "Core plan $29 incl. 100k units, +$8 per extra 100k for all extra units (vendor says overage 'from $8', so check its tiered rates at high volume). Assumes 1 unit per request; real traces can have several observations."
      },
      "news": [
        {
          "value": "ClickHouse acquired Langfuse (announced 16 Jan 2026); Langfuse stays open source.",
          "source": "https://clickhouse.com/blog/clickhouse-acquires-langfuse-open-source-llm-observability"
        }
      ],
      "pricing_public": {
        "value": "yes",
        "source": "https://langfuse.com/pricing"
      },
      "score_points": 6,
      "score_max": 8,
      "score_pct": 75,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Traceloop / OpenLLMetry",
      "slug": "traceloop",
      "vendor": "Traceloop (ServiceNow)",
      "url": "https://www.traceloop.com",
      "category": "Observability",
      "summary": "OpenTelemetry-based LLM instrumentation (OpenLLMetry) plus a hosted observability platform; cost estimates as OTel attributes.",
      "open_source": {
        "value": "yes",
        "detail": "OpenLLMetry SDK is Apache-2.0; hosted platform is commercial",
        "source": "https://github.com/traceloop/openllmetry"
      },
      "hosting": {
        "value": "Both",
        "detail": "Hosted platform; SDK exports to any OTel backend",
        "source": "https://www.traceloop.com/pricing"
      },
      "hard_budgets": {
        "value": "n/a",
        "detail": "Not in the request path",
        "source": null
      },
      "rate_limits": {
        "value": "n/a",
        "source": null
      },
      "cost_attribution": {
        "value": "partial",
        "detail": "Token usage and cost estimates as OpenTelemetry attributes",
        "source": "https://www.traceloop.com/blog/visualizing-llm-performance-with-opentelemetry-tools-for-tracing-cost-and-latency"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "n/a",
        "detail": "Instrumentation, not a proxy",
        "source": null
      },
      "caching": {
        "value": "n/a",
        "source": null
      },
      "eu_residency": {
        "value": "unknown",
        "source": null
      },
      "pricing_model": {
        "value": "Free tier + enterprise quote",
        "source": "https://www.traceloop.com/pricing"
      },
      "price_points": {
        "value": "Free up to 50k spans/mo; Enterprise quote (AWS Marketplace lists $15,000/yr)",
        "source": "https://aws.amazon.com/marketplace/pp/prodview-oyluqesay6yoy"
      },
      "calc": {
        "type": "custom"
      },
      "news": [
        {
          "value": "Traceloop is joining ServiceNow (2026); OpenLLMetry stays open source.",
          "source": "https://www.traceloop.com/blog/traceloop-is-joining-servicenow"
        }
      ],
      "pricing_public": {
        "value": "partial",
        "source": "https://aws.amazon.com/marketplace/pp/prodview-oyluqesay6yoy"
      },
      "score_points": 2,
      "score_max": 8,
      "score_pct": 25,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "LangSmith (incl. LLM Gateway)",
      "slug": "langsmith",
      "vendor": "LangChain",
      "url": "https://www.langchain.com/langsmith",
      "category": "Observability",
      "summary": "Tracing/evals platform that now includes an LLM Gateway with hard USD spend caps per org, workspace, key or user.",
      "open_source": {
        "value": "no",
        "source": "https://www.langchain.com/pricing"
      },
      "hosting": {
        "value": "Both",
        "detail": "Cloud (US or EU); hybrid and self-hosted on Enterprise",
        "source": "https://www.langchain.com/pricing"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Hard USD caps per organization, workspace, API key or user; hourly to monthly; HTTP 402",
        "source": "https://docs.langchain.com/langsmith/llm-gateway-spend-policies"
      },
      "rate_limits": {
        "value": "unknown",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Cost per trace, project and dashboard; per-customer header policies",
        "source": "https://docs.langchain.com/langsmith/cost-tracking"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "unknown",
        "source": null
      },
      "caching": {
        "value": "unknown",
        "source": null
      },
      "eu_residency": {
        "value": "yes",
        "detail": "LangChain cloud in US or EU",
        "source": "https://www.langchain.com/pricing"
      },
      "pricing_model": {
        "value": "Per seat + usage",
        "source": "https://www.langchain.com/pricing"
      },
      "price_points": {
        "value": "Developer $0 (1 seat, 5k base traces); Plus $39/seat/mo (10k base traces) then pay-as-you-go; Enterprise custom",
        "source": "https://www.langchain.com/pricing"
      },
      "calc": {
        "type": "partial",
        "base": 39,
        "note": "Plus: $39 per seat; trace overage rates not shown in page text."
      },
      "news": [],
      "pricing_public": {
        "value": "partial",
        "source": "https://www.langchain.com/pricing"
      },
      "score_points": 7,
      "score_max": 12,
      "score_pct": 58,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Vantage",
      "slug": "vantage",
      "vendor": "Vantage",
      "url": "https://www.vantage.sh",
      "category": "FinOps platform",
      "summary": "Cloud cost platform with OpenAI/Anthropic connectors, virtual tagging and FOCUS 1.3 import/export; public tiered pricing.",
      "open_source": {
        "value": "no",
        "source": "https://www.vantage.sh/pricing"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://www.vantage.sh/pricing"
      },
      "hard_budgets": {
        "value": "n/a",
        "detail": "Reporting platform, not in the request path",
        "source": null
      },
      "rate_limits": {
        "value": "n/a",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Virtual tagging (Business and up)",
        "source": "https://www.vantage.sh/pricing"
      },
      "focus_export": {
        "value": "yes",
        "detail": "FOCUS 1.3 imports and exports",
        "source": "https://www.vantage.sh/blog/focus-1-3-support"
      },
      "focus_version": "1.3",
      "providers": {
        "value": "30+ cost providers",
        "detail": "Includes Anthropic (https://www.vantage.sh/blog/anthropic-support)",
        "source": "https://www.vantage.sh/pricing"
      },
      "caching": {
        "value": "n/a",
        "source": null
      },
      "eu_residency": {
        "value": "unknown",
        "source": null
      },
      "pricing_model": {
        "value": "Tiered by tracked spend",
        "source": "https://www.vantage.sh/pricing"
      },
      "price_points": {
        "value": "Starter free (≤$2,500 tracked spend/mo); Pro $30/mo (≤$7,500); Business $200/mo (≤$20,000); Enterprise custom (% of spend)",
        "source": "https://www.vantage.sh/pricing"
      },
      "calc": {
        "type": "vantage",
        "note": "Tier chosen by your estimated monthly AI spend (tokens × price). Assumes AI is the only spend tracked."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://www.vantage.sh/pricing"
      },
      "score_points": 6,
      "score_max": 8,
      "score_pct": 75,
      "providers_num": 30,
      "last_checked": "2026-10-20"
    },
    {
      "name": "CloudZero",
      "slug": "cloudzero",
      "vendor": "CloudZero",
      "url": "https://www.cloudzero.com",
      "category": "FinOps platform",
      "summary": "Cost intelligence platform with native OpenAI/Anthropic connections and AnyCost ingestion for any source.",
      "open_source": {
        "value": "no",
        "source": "https://www.cloudzero.com/integrations/"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://www.cloudzero.com/integrations/"
      },
      "hard_budgets": {
        "value": "n/a",
        "detail": "Not in the request path",
        "source": null
      },
      "rate_limits": {
        "value": "n/a",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "OpenAI by project/model/service tier; Anthropic by project/model; Claude Enterprise per user",
        "source": "https://docs.cloudzero.com/docs/ai-platforms"
      },
      "focus_export": {
        "value": "partial",
        "detail": "FOCUS import (FOCUS Adaptor); native FOCUS export not documented",
        "source": "https://www.cloudzero.com/press-releases/20240620/"
      },
      "focus_version": "import only",
      "providers": {
        "value": "unknown",
        "source": null
      },
      "caching": {
        "value": "n/a",
        "source": null
      },
      "eu_residency": {
        "value": "unknown",
        "source": null
      },
      "pricing_model": {
        "value": "unknown",
        "detail": "Not public",
        "source": null
      },
      "price_points": {
        "value": "unknown",
        "detail": "Not public",
        "source": null
      },
      "calc": {
        "type": "custom"
      },
      "news": [],
      "pricing_public": {
        "value": "no",
        "source": null
      },
      "score_points": 3,
      "score_max": 8,
      "score_pct": 38,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Finout",
      "slug": "finout",
      "vendor": "Finout",
      "url": "https://www.finout.io",
      "category": "FinOps platform",
      "summary": "FinOps platform with AI virtual tags for OpenAI/Anthropic spend allocation and FOCUS-based dimensions.",
      "open_source": {
        "value": "no",
        "source": "https://www.finout.io/pricing"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://www.finout.io/pricing"
      },
      "hard_budgets": {
        "value": "n/a",
        "detail": "Not in the request path",
        "source": null
      },
      "rate_limits": {
        "value": "n/a",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Virtual tags by team, product, model, customer or agent",
        "source": "https://www.finout.io/megabill/ai-virtual-tag"
      },
      "focus_export": {
        "value": "partial",
        "detail": "FOCUS used as a data model / virtual tags; export format not confirmed",
        "source": "https://docs.finout.io/get-started-with-finout/cost-and-usage-types/focus-in-finout"
      },
      "focus_version": "unknown",
      "providers": {
        "value": "unknown",
        "source": null
      },
      "caching": {
        "value": "n/a",
        "source": null
      },
      "eu_residency": {
        "value": "unknown",
        "source": null
      },
      "pricing_model": {
        "value": "Annual subscription by committed spend (quote)",
        "source": "https://www.finout.io/pricing"
      },
      "price_points": {
        "value": "Not public",
        "source": "https://www.finout.io/pricing"
      },
      "calc": {
        "type": "custom"
      },
      "news": [],
      "pricing_public": {
        "value": "no",
        "source": "https://www.finout.io/pricing"
      },
      "score_points": 3,
      "score_max": 8,
      "score_pct": 38,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Pay-i",
      "slug": "pay-i",
      "vendor": "Pay-i",
      "url": "https://pay-i.com",
      "category": "FinOps platform",
      "summary": "GenAI cost, capacity and ROI platform with Allow/Block limits at any scope (user, team, use case).",
      "open_source": {
        "value": "no",
        "source": "https://pay-i.com/products/roi-optimization-platform"
      },
      "hosting": {
        "value": "unknown",
        "source": null
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Limits in Allow (track) or Block (halt proxied requests) mode",
        "source": "https://docs.pay-i.com/docs/budgets-2"
      },
      "rate_limits": {
        "value": "unknown",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Spend by user, team, application, use case; linked to business KPIs",
        "source": "https://pay-i.com/products/roi-optimization-platform"
      },
      "focus_export": {
        "value": "unknown",
        "detail": "No FOCUS export found",
        "source": null
      },
      "focus_version": null,
      "providers": {
        "value": "unknown",
        "source": null
      },
      "caching": {
        "value": "n/a",
        "source": null
      },
      "eu_residency": {
        "value": "unknown",
        "source": null
      },
      "pricing_model": {
        "value": "unknown",
        "detail": "Not public",
        "source": null
      },
      "price_points": {
        "value": "unknown",
        "detail": "Not public",
        "source": null
      },
      "calc": {
        "type": "custom"
      },
      "news": [],
      "pricing_public": {
        "value": "no",
        "source": null
      },
      "score_points": 4,
      "score_max": 12,
      "score_pct": 33,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Revenium",
      "slug": "revenium",
      "vendor": "Revenium",
      "url": "https://www.revenium.ai",
      "category": "FinOps platform",
      "summary": "AI cost metering with attribution to customers/features, hard budget controls and a FOCUS 1.0 export.",
      "open_source": {
        "value": "no",
        "source": "https://www.revenium.ai/pricing"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://www.revenium.ai/pricing"
      },
      "hard_budgets": {
        "value": "yes",
        "detail": "Hard budget controls per team, customer or feature (higher tiers); budgets & alerts docs: https://docs.revenium.io/track-and-control-costs/set-budgets-and-alerts",
        "source": "https://www.revenium.ai/pricing"
      },
      "rate_limits": {
        "value": "yes",
        "detail": "Policy enforcement: block or throttle requests above cost thresholds",
        "source": "https://www.revenium.ai/pricing"
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Cost per customer, feature, workflow, team",
        "source": "https://www.revenium.ai/pricing"
      },
      "focus_export": {
        "value": "yes",
        "detail": "FOCUS v1.0 export added in v2.6.0 (19 Feb 2026); CSV/Parquet to S3 or Snowflake",
        "source": "https://docs.revenium.io/get-started/changelog"
      },
      "focus_version": "1.0",
      "providers": {
        "value": "5+ named + custom",
        "detail": "OpenAI, Anthropic, Azure OpenAI, AWS Bedrock, Google Vertex AI, custom via SDK",
        "source": "https://www.revenium.ai/pricing"
      },
      "caching": {
        "value": "n/a",
        "source": null
      },
      "eu_residency": {
        "value": "unknown",
        "source": null
      },
      "pricing_model": {
        "value": "Tiered by transactions",
        "source": "https://www.revenium.ai/pricing"
      },
      "price_points": {
        "value": "Free 100k transactions/mo; Growth 350k/mo (base price not shown), overage $750 per 500k ($1.50/1k); Enterprise 1M+ custom",
        "source": "https://www.revenium.ai/pricing"
      },
      "calc": {
        "type": "revenium",
        "note": "Free up to 100k transactions (1 per tracked request). Growth base price is not public, so above 100k only the overage component can be shown."
      },
      "news": [],
      "pricing_public": {
        "value": "partial",
        "source": "https://www.revenium.ai/pricing"
      },
      "score_points": 9,
      "score_max": 12,
      "score_pct": 75,
      "providers_num": 5,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Ternary",
      "slug": "ternary",
      "vendor": "Ternary",
      "url": "https://ternary.app",
      "category": "FinOps platform",
      "summary": "Multicloud FinOps platform that ingests AI gateway spend (e.g. LiteLLM) as FOCUS 1.2 for allocation.",
      "open_source": {
        "value": "no",
        "source": "https://aws.amazon.com/marketplace/pp/prodview-mp2iwlzw2a6n6"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://aws.amazon.com/marketplace/pp/prodview-mp2iwlzw2a6n6"
      },
      "hard_budgets": {
        "value": "n/a",
        "detail": "Not in the request path",
        "source": null
      },
      "rate_limits": {
        "value": "n/a",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Split shared AI invoices across teams, projects and API keys",
        "source": "https://docs.litellm.ai/docs/observability/ternary"
      },
      "focus_export": {
        "value": "partial",
        "detail": "Ingests FOCUS (AI gateway connections take FOCUS 1.2 CSV) and documents FOCUS mappings; a FOCUS export was not confirmed",
        "source": "https://docs.ternary.app/docs/ternary-mappings-to-focus-mappings"
      },
      "focus_version": "1.2 (ingest)",
      "providers": {
        "value": "unknown",
        "source": null
      },
      "caching": {
        "value": "n/a",
        "source": null
      },
      "eu_residency": {
        "value": "unknown",
        "source": null
      },
      "pricing_model": {
        "value": "Annual, by managed spend",
        "source": "https://aws.amazon.com/marketplace/pp/prodview-mp2iwlzw2a6n6"
      },
      "price_points": {
        "value": "AWS Marketplace: $25,000/yr up to $1M managed spend; $67,500 up to $3M; $100,000 up to $5M; $175,000 above",
        "source": "https://aws.amazon.com/marketplace/pp/prodview-mp2iwlzw2a6n6"
      },
      "calc": {
        "type": "ternary",
        "note": "AWS Marketplace annual tiers by managed spend ($25k up to $1M/yr, $67.5k up to $3M, $100k up to $5M, $175k above). Assumes only AI spend is managed."
      },
      "news": [],
      "pricing_public": {
        "value": "yes",
        "source": "https://aws.amazon.com/marketplace/pp/prodview-mp2iwlzw2a6n6"
      },
      "score_points": 5,
      "score_max": 8,
      "score_pct": 62,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    },
    {
      "name": "Flexera One (AI cost management)",
      "slug": "flexera-one",
      "vendor": "Flexera",
      "url": "https://www.flexera.com/products/flexera-one/ai-cost-management",
      "category": "FinOps platform",
      "summary": "Enterprise FinOps suite with OpenAI/Anthropic cost tracking, allocation and budgets; ingests FOCUS bill data.",
      "open_source": {
        "value": "no",
        "source": "https://www.flexera.com/products/flexera-one/ai-cost-management"
      },
      "hosting": {
        "value": "SaaS",
        "source": "https://www.flexera.com/products/flexera-one/ai-cost-management"
      },
      "hard_budgets": {
        "value": "n/a",
        "detail": "Budgets/alerts on reported spend; not in the request path",
        "source": null
      },
      "rate_limits": {
        "value": "n/a",
        "source": null
      },
      "cost_attribution": {
        "value": "yes",
        "detail": "Allocation of model/token spend",
        "source": "https://www.flexera.com/products/flexera-one/ai-cost-management"
      },
      "focus_export": {
        "value": "partial",
        "detail": "Ingests FOCUS 1.0 bill data (with selected 1.2 fields); export not confirmed",
        "source": "https://docs.flexera.com/flexera-one/administration/cloud-settings/bill-data-connections/bill-connect-configurations/finops-open-cost-and-usage-specification-focus/"
      },
      "focus_version": "1.0 (+ some 1.2 fields), ingest",
      "providers": {
        "value": "unknown",
        "source": null
      },
      "caching": {
        "value": "n/a",
        "source": null
      },
      "eu_residency": {
        "value": "unknown",
        "source": null
      },
      "pricing_model": {
        "value": "unknown",
        "detail": "Not public",
        "source": null
      },
      "price_points": {
        "value": "unknown",
        "detail": "Not public",
        "source": null
      },
      "calc": {
        "type": "custom"
      },
      "news": [],
      "pricing_public": {
        "value": "no",
        "source": null
      },
      "score_points": 3,
      "score_max": 8,
      "score_pct": 38,
      "providers_num": -1,
      "last_checked": "2026-10-20"
    }
  ]
}