OpenRouter vs Standard Compute
OpenRouter bills every token at the provider's price plus a 5.5% credit fee, so an agent's month costs whatever it ran; Standard Compute puts the compute inside a flat monthly plan and routes routine steps to cost-efficient models, which makes it the better fit for coding agents that run every day.
Side by side
| OpenRouter | Standard Compute | |
|---|---|---|
| Pricing | Provider price per token + 5.5% on credit purchases ($0.80 minimum); 8% on Business | Flat monthly plan: $19, $39, $89 or $249 with $20 to $269 of compute included; Business from $499 |
| Who picks the model | You pin a model, or use openrouter/auto or openrouter/pareto-code | Router picks per request (harder steps to stronger models, routine steps to efficient ones), or pin any model ID |
| Budget control | Credit limit per key with daily, weekly or monthly reset; guardrail budgets | Plan budget per key; new requests get HTTP 402 when it is used, no overage |
| Claude Code | Native Messages API | Native Messages API |
| Codex | Native Responses API | Native Responses API, any model in the catalog |
| Europe / US rules | EU and US in-region routing on Business and Enterprise | Opt-in Europe or US rules on every plan, per key, team or company |
| Teams | Workspaces with member and key guardrails | Company and agency workspaces, a plan per key |
| Best for | Model breadth, provider pinning, light or irregular use | Coding agents and agent fleets that need one fixed monthly cost |
OpenRouter facts from its own pricing and docs pages. Standard Compute is our product.
Where they differ
- OpenRouter is provider price per token plus 5.5% when you buy credits (8% on Business). Standard Compute is $19 to $249 a month with $20 to $269 of compute included, and new requests stop with HTTP 402 when it is used.
- OpenRouter's Auto Router ranks models by what its community spends on for each task type, and its Pareto router applies one coding-score floor to every request. Standard Compute's router decides per request, sending harder steps to stronger models and routine steps to efficient ones.
- OpenRouter's EU and US in-region routing needs Business or Enterprise. Standard Compute's Europe or US rules are opt-in on every plan, and company workspaces give each key its own plan.
When OpenRouter is the better fit
OpenRouter is the better pick when you need one of its 500+ models or a specific provider, want to bring your own keys, or use an agent lightly enough that per-token billing costs less than a plan.
Switch from OpenRouter 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 | OpenRouter | Standard Compute |
|---|---|---|
| OpenAI-compatible base URL | https://openrouter.ai/api/v1 | https://api.stdcmpt.com/v1 |
| Claude Code base URL | https://openrouter.ai/api | https://api.stdcmpt.com |
| API key | Your OpenRouter 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"Run your agents on a fixed monthly budget
Every plan includes the compute and every model, and the router spends the budget on the most cost-efficient model for each step. Pay-as-you-go top-ups are optional, never automatic.
FAQ
Is Standard Compute cheaper than OpenRouter?
Pinned to one model, the cost is close: OpenRouter adds 5.5% to provider prices, and a Standard Compute plan includes 5% to 14% more compute than its price if you use it. Standard Compute saves money by routing routine steps to efficient models; on a median coding day in our Agent Burn Index those cost about 15 times less than Claude Sonnet 5.5. Light or irregular use is cheaper on OpenRouter.
Does OpenRouter work with Claude Code?
Yes. OpenRouter serves the Anthropic Messages API, so Claude Code connects with ANTHROPIC_BASE_URL=https://openrouter.ai/api and your OpenRouter key. Standard Compute works the same way with ANTHROPIC_BASE_URL=https://api.stdcmpt.com.
Which has more models?
OpenRouter, by far: 500+ models from 80+ providers. Standard Compute carries 35 models from 11 makers (OpenAI, Anthropic, Google, Qwen, DeepSeek, MiniMax, Z.AI, Moonshot, Xiaomi, xAI, Mistral), and every paid plan can use all of them.
What happens when the money runs out?
On OpenRouter, a key that reaches its credit limit is refused until the limit resets, and an account without credits gets HTTP 402. On Standard Compute, new requests get HTTP 402 with the code budget_exhausted when the plan budget is used; requests already running finish, and nothing is billed past the plan unless you turn on pay-as-you-go top-ups.
Does Standard Compute add a fee on top of model prices?
No. Usage counts against the plan budget at the model's per-token rate and the plan price is all you pay. Optional pay-as-you-go top-ups, on Economy and above, add compute at the usage rate plus 15%.
Sources: OpenRouter pricing · OpenRouter FAQ · OpenRouter in-region routing · OpenRouter Claude Code