Portkey vs Standard Compute
Portkey, now Palo Alto Networks' Prisma AIRS AI Gateway, charges for logged requests on top of your own provider bills, while Standard Compute optimizes the bill itself: its router picks the cheapest model that can do each request, with the compute included in one plan, which suits a small team or agency that wants agents running at lower cost.
Side by side
| Portkey | Standard Compute | |
|---|---|---|
| Who picks the model | Configs you write: fallbacks, load balancing, conditions | Router picks the cheapest model that can do each request (harder steps to stronger models), or pin any model ID |
| Pricing | Developer free (10k logs); Production $49/mo for 100k logs, $9 per extra 100k; you pay providers | Fixed monthly plan: $19, $39, $89 or $249 with compute equal to the price included; Business from $499 |
| Budget control | USD or token limits per integration and workspace | Plan budget per key; new requests get HTTP 402 when it is used, no overage |
| Claude Code | Native | Native Messages API |
| Codex | Documented | Native Responses API, any model in the catalog |
| Europe / US rules | Hybrid data plane in your own cloud on Enterprise | Opt-in Europe or US rules on every plan, per key, team or company |
| Teams | Workspaces; SSO and RBAC on Enterprise | Company and agency workspaces, a plan per key |
| Best for | Enterprises standardizing on Palo Alto security | Coding agents and fleets that should cost less per task, at one fixed monthly price |
Portkey facts from its own pricing and docs pages. Standard Compute is our product.
Where they differ
- Portkey's Production plan is $49 a month for 100,000 logged requests plus $9 per extra 100,000, and you bring and pay your own provider keys. Standard Compute's plan includes the compute.
- Portkey runs fallback, load-balancing and conditional configs that you write. Standard Compute picks the model per request.
- Portkey brings enterprise security tooling: guardrails, audit logs and a hybrid data plane in your own cloud. Standard Compute is a hosted endpoint with per-key plans and region rules, without that tooling.
When Portkey is the better fit
Portkey is the better pick for an enterprise that needs a data plane in its own cloud, security-suite governance or a Palo Alto standard.
Switch from Portkey in 2 minutes
Standard Compute speaks the same APIs, so switching is a base URL, a key and a model name. Your prompts, tools and streaming stay as they are.
| Setting | Portkey | Standard Compute |
|---|---|---|
| OpenAI-compatible base URL | https://api.portkey.ai/v1 | https://api.stdcmpt.com/v1 |
| Claude Code base URL | https://api.portkey.ai | https://api.stdcmpt.com |
| API key | Your Portkey key | Your Standard Compute key |
| Model | The model ID you use today | standardcompute, or any exact model ID |
# OpenAI-compatible clients (OpenCode, Cline, Kilo Code, SDKs)
base_url = "https://api.stdcmpt.com/v1"
api_key = "<your-standard-compute-key>"
model = "standardcompute" # or pin one, e.g. anthropic/claude-sonnet-5.5# Claude Code (Anthropic Messages API, no /v1)
export ANTHROPIC_BASE_URL="https://api.stdcmpt.com"
export ANTHROPIC_AUTH_TOKEN="<your-standard-compute-key>"# Codex CLI, ~/.codex/config.toml
model = "standardcompute"
model_provider = "standardcompute"
[model_providers.standardcompute]
name = "Standard Compute"
base_url = "https://api.stdcmpt.com/v1"
env_key = "STANDARD_COMPUTE_API_KEY"
wire_api = "responses"More intelligence per dollar
The router sends each step to the cheapest model that can do it, and for every model we vet providers for price, uptime and speed. Europe or US rules are opt-in, we don't train on your prompts, and every model comes on one fixed monthly plan. Pay-as-you-go top-ups are optional, never automatic.
FAQ
Is Portkey part of Palo Alto Networks now?
Yes. Palo Alto Networks completed its acquisition of Portkey on May 29, 2026, and Portkey's docs now call the product Prisma AIRS AI Gateway.
How much does Portkey cost?
Portkey's pricing page lists a free Developer tier with 10,000 logs a month and a Production plan at $49 a month for 100,000 logs plus $9 per extra 100,000; Enterprise is custom. Model usage is billed by your providers on your own keys.
Does Portkey pick the cheapest model?
Not automatically. Portkey routes with the fallback, load-balancing and conditional configs you write. Standard Compute's router picks a cost-efficient model per request.
Does Portkey work with Claude Code?
Yes. Portkey documents Claude Code with ANTHROPIC_BASE_URL=https://api.portkey.ai. Standard Compute works with ANTHROPIC_BASE_URL=https://api.stdcmpt.com.
Who should choose Standard Compute over Portkey?
A small team or agency that wants one bill with compute included, a budget per key and automatic routing, and has no need for an enterprise security suite or a data plane in its own cloud.
Sources: Portkey pricing · Portkey docs (llms.txt) · Portkey with Claude Code