Pricing reference
The full breakdown. For the sales pitch, see the pricing page.
Tier comparison
| Free | Team | Business | Scale | Enterprise | |
|---|---|---|---|---|---|
| Price | $0 | $99/mo | $499/mo | $1,499/mo | Custom |
| BYOK fair-use | 100,000 BYOK req / mo | 5,000,000 BYOK req / mo | Unlimited BYOK | Unlimited BYOK | Custom |
| Managed-key tokens | $5 trial credit · PAYG add-on | Provider cost + 4% gateway fee | Provider cost + 4% gateway fee | Provider cost + 4% gateway fee | Custom |
| Rate limit | 120 RPM | 1,500 RPM | 10,000 RPM | 30,000 RPM | Custom |
| Receipt retention | 7-day routing data | 30-day routing data | 180-day routing data | 12-month routing data | Custom |
| Managed-key fee | Trial credit (no fee) | 4% platform fee on managed keys | 4% platform fee on managed keys | 4% platform fee on managed keys | Custom |
The 4% platform fee applies only to managed-key traffic and covers unified billing, routing, and key rotation. On BYOK, providers bill you directly at zero markup; the per-month request quota is a fair-use guardrail, not a metering scheme.
Provider rates
What you pay the provider, per 1M tokens. Synced daily.
| Model | Provider | Input / 1M | Output / 1M |
|---|---|---|---|
| GPT-5.4 | OpenAI | $2.50 | $15.00 |
| Claude Opus 4.7 | Anthropic | $5.00 | $25.00 |
| Claude Opus 4.6 | Anthropic | $5.00 | $25.00 |
| Gemini 3.1 Pro | $2.00 | $12.00 | |
| Grok 4.1 Fast | xAI | $3.00 | $15.00 |
| GPT-5.4 Mini | OpenAI | $0.75 | $4.50 |
| Claude Sonnet 4.6 | Anthropic | $3.00 | $15.00 |
| DeepSeek R2 | DeepSeek | $0.30 | $1.20 |
| GPT-5.2 | OpenAI | $1.75 | $14.00 |
| GPT-4.1 | OpenAI | $2.00 | $8.00 |
| Gemini 3 Flash | $0.50 | $3.00 | |
| GPT-5.3 Codex | OpenAI | $2.00 | $10.00 |
| Qwen3 Coder 480B (Together) | Together AI | $0.20 | $0.60 |
| DeepSeek V3.2 | DeepSeek | $0.14 | $0.28 |
| MiniMax M2.5 | MiniMax | $0.15 | $0.60 |
| Claude Haiku 4.5 | Anthropic | $1.00 | $5.00 |
| GPT-4.1 Mini | OpenAI | $0.40 | $1.60 |
| Mistral Large 3 | Mistral | $0.50 | $1.50 |
| Grok 4.1 Mini | xAI | $0.30 | $1.50 |
| GPT-5.4 Nano | OpenAI | $0.20 | $1.25 |
| Gemini 3.1 Flash Lite | $0.25 | $1.50 | |
| Gemini 2.5 Flash | $0.30 | $2.50 | |
| GPT-4.1 Nano | OpenAI | $0.10 | $0.40 |
| Llama 3.3 70B Versatile | Groq | $0.59 | $0.79 |
| auto | Auto-routed | Cheapest model that clears your quality bar | |
Catalog last synced: Apr 20, 2026, 08:00 UTC · 48 models, 11 providers
Metering rules
Managed-key input + output tokens
Provider cost + 4% gateway fee. No allotment cap, no overage.
Fallback hops
If a primary errors and we retry, both attempts meter (each at provider cost + 4%).
BYOK requests
Counted against the per-month fair-use quota on Free and Team. Business and Scale: unlimited. Your provider bill stays with the provider at zero markup.
Filtered candidates
Models the router rejects pre-dispatch don’t meter.
Cache hits
Prompt-cache hits are free.
FAQ
How does managed-key billing work?
Provider cost passthrough plus a flat 4% gateway fee. There is no token allotment and no overage rate — usage scales linearly with traffic, and the 4% scales with it. Drawn from a prepaid balance; auto-replenish on a threshold is optional.
What happens if I run hot one month?
Nothing special — there is no surprise overage line. Managed-key usage shows up on the receipt as provider cost + 4%, and you pay what you actually used. Hard caps still exist as an opt-in toggle on any API key for a per-key spend ceiling.
What is the BYOK quota for?
On Free and Team, BYOK is capped at a per-month request count for fair-use protection (the routing, receipts, and dashboard infrastructure isn’t free to operate). Business and Scale are unlimited. The quota guardrails the platform; it is not a metering scheme.
BYOK on every tier?
Yes, including Free. Full routing, receipts, dashboard, and OTel export either way.
Start free.
100K BYOK requests / month plus a $5 managed-key trial credit. No card.
Get your API key →