Monthly Tokens
Route and report high-volume text, image, and video usage through one Quotaflow control layer.
AI Gateway
The wallet execution layer for every AI call.
Run OpenAI-compatible requests through one AI token wallet for BYOK keys, approved supply, route policy, token ledger, AI spend management, and ROI evidence.
chat.completions
POST /v1/chat/completions
92 ms route decision
model openai:gpt-5.4
wallet qf_main
ledger usage+roiWhy Quotaflow
AI Gateway for LLM cost optimization.
Lower-cost approved routes across leading models, wallet-scoped BYOK policy controls without vendor lock-in, and a token ledger that connects every API call back to AI spend ROI.
Global Partners
Provider and capacity relationships for lower-cost, policy-approved routes worldwide.
Monthly transactions
Monthly AI transaction volume tracked across routing, quota, audit, and reporting workflows.
On one endpoint
GPT, Claude, Gemini, embeddings, image, and video routes behind a single API and reporting layer.
Flexible token access
Choose how the wallet spends.
Use the wallet in two official-provider modes: partner-routed BYOK capacity or dedicated provider key paths. Choose the shape that fits security, fund safety, quality requirements, cashback eligibility, and route provenance.
Routing partners provide official keys from major provider channels.
AWSpartner key sourceCloud-provider capacity
Azure OpenAIpartner key sourceMicrosoft-hosted OpenAI
OpenAIpartner key sourceOfficial OpenAI account paths
Anthropicpartner key sourceOfficial Claude account paths
GCPpartner key sourceGoogle Cloud and Gemini access
Vercelpartner key sourceApp and AI gateway channels
ByteDancepartner key sourceSeedance and model capacity
And morepartner key sourceAdditional verified partnersDedicated mode pins traffic to a provider-owned key path for stricter control.
AWS Claudededicated key pathClaude through AWS capacityAnthropic directdedicated key pathOfficial Anthropic account pathAzure OpenAIdedicated key pathDedicated Azure OpenAI accountOpenAI directdedicated key pathOfficial OpenAI account pathGCP Geminidedicated key pathDedicated Google Cloud accessAnd morededicated key pathVerified official capacity optionsHow it works
Choose how you buy and use tokens.
Start with flexible routing for pay-as-you-go usage, or choose a dedicated provider path when security policy, fund safety, or quality requirements need official terms and prepaid capacity.
Routing mode
Choose routing mode when you want partner-provided official keys, elastic capacity, and pay-as-you-go token access without committing to a provider prepay schedule.
Purchase flow
- 1. Choose routing
partner-routed BYOKMODEPick routing when you want Quotaflow to select from approved partner key paths while keeping provenance visible.
View pricing - 2. Add payment method
pay as you goPAYGNo provider minimum is required. Usage is metered as requests run through the gateway.
View pricing - 3. Get your API key
OpenAI-compatibleAPIUse one Quotaflow API key and keep existing SDK request shapes for GPT, Claude, Gemini, image, and video routes.
View pricing
Reliability layer
Know the RPM, failover, and token record behind every model.
Quotaflow keeps each model route operationally transparent: target RPM, live availability, failover options, and detailed token usage records for every request.
GPT-5.5openai:gpt-5.5Frontier route with RPM targets, availability tracking, and auditable token records.
Claude Opus 4.8anthropic:claude-opus-4-8Dedicated and routed Claude paths with official-provider failover options.
Gemini 3.5 Flashgoogle:gemini-3.5-flashFast multimodal route monitored for throughput, latency, SSE health, and quota headroom.
GPT-5.4 Miniopenai:gpt-5.4-miniLower-latency route with transparent per-request token usage and workspace reporting.
Target RPMFailover optionsLive healthToken ledgerPer modelPer provider pathPer workspacePer request
LLM API pricing
Full LLM pricing What you would save by routing with Quotaflow.
Pick a workload and current monthly AI spend. Quotaflow maps your provider bill to approved lower-cost routes and shows the monthly saving at your scale.
Use case
Comparison against public provider baselines using estimated Quotaflow pricing where supported.
OpenAI-compatible LLM API
Already on the OpenAI SDK? Change two values.
The Quotaflow text endpoint keeps the familiar request shape. Point your client at Quotaflow and attach metadata for reports.
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.QUOTAFLOW_API_KEY,
baseURL: "https://api.quotaflow.ai/openai/v1"
});
await client.responses.create({
model: "openai:gpt-5.4",
input: "Refactor this function",
metadata: { team: "engineering" }
});curl https://api.quotaflow.ai/openai/v1/models -H "Authorization: Bearer $QUOTAFLOW_API_KEY"
# Use the exact model id returned by /models.
# Unsupported ids fail fast instead of being swapped.Built for production
LLM inference built for demanding products.
OpenAI-compatible LLM API
Keep existing SDK calls, change the base URL and key, then attach metadata for reporting.
Claude-compatible Messages
Claude-style clients can use exact enabled Claude model ids returned by discovery.
Gemini-compatible generateContent
Gemini-native clients can keep countTokens and cached-content workflows where supported.
Shared usage reporting
Track route, owner, workflow, and estimated savings from one shared balance.
Endpoint discovery
Use /models to verify exactly which model ids are enabled for a key.
Talk to us for scale
Scope private route review, procurement mapping, and committed-use reporting before rollout.
Security
Your data stays yours.
FAQ
Get an API key.
Start with routing mode, or talk to us for a dedicated provider path with prepaid capacity and transparent token reconciliation.
Official provider routes, failover visibility, and token ledger in one API.
Get started
