Spend Less on AI
Keep the Quality
Batch pricing around 40% below realtime API rates on most supported models. No queues to manage, no minimum volumes — just send requests and get results.
# Load your cargo
curl -X POST \
https://api.cnvy.ai/cargo/load \
-H "Content-Type: application/json" \
-H "X-API-Key: $API_KEY" \
-d '{"params": {"model": "convoy-mock",
"messages": [{"role": "user", "content": "Hello"}]},
"callback_url": "https://example.com/callback"}'Why Choose Convoy
Infrastructure that gets out of your way so you can focus on building.
Cost Savings
Take advantage of batch pricing without infrastructure complexity — around 40% below realtime API rates on most supported models.
Zero Batching Logic
No queues to manage, no timing windows to configure — just send requests and we handle the rest.
Multimodal Built In
Upload images, PDFs, and documents once, then reference them across thousands of batch requests. Every file is malware-scanned before use.
No Minimum Volume
Start with one request or send thousands. Convoy scales seamlessly with your workload.
Built-in Reliability
Automatic retry logic, error handling, and delivery tracking help make sure results reach your callback.
Leading models through one API.
Access Claude, Llama, Mistral, Nova, and more — all through a single API. No vendor lock-in. Pick a different model for every request — no code changes, no extra infrastructure.
Every industry has a batch AI backlog.
High-volume, latency-tolerant workloads — text and multimodal — that are perfectly suited for async batch processing, across every vertical.
Healthcare & Medical
Radiology report drafting, clinical note summarization, ICD coding, and prior authorization — including scanned reports and imaging attachments, processed overnight as batch jobs.
Legal & Compliance
Discovery document review, M&A due diligence, and PDF contract portfolio abstraction — thousands of documents processed in hours.
Financial Services
AML narratives, credit memo generation, earnings call analysis, and wealth reporting — overnight data pipelines native to finance.
Voice, Audio & Video
Call center QA at 100% volume, sales call intelligence, meeting summaries, and media post-production — all queued and processed async.
Document Processing
Invoice extraction, form digitization, archive classification — PDFs, scans, and images, the most common enterprise AI use case, delivered in hours instead of months.
Marketing & Content
Weekly ad copy variants, email campaign drafts, and content calendars generated in overnight batches — consistent brand voice at scale.
Connects to your entire stack
From Python scripts to no-code platforms — Convoy works wherever you do. Send requests from any tool that speaks HTTP.
Zapier
Connect to 6,000+ apps
Make
Visual workflows with branching
Retool
Internal AI tools & dashboards
Slack
Team AI assistant
Google Sheets
Spreadsheet-powered AI
Webhooks
Universal callback pattern
Python
Scripts, apps & data pipelines
Node.js
Web apps & APIs
AWS Lambda
Serverless event-driven pipelines
LangChain
Batch AI with LLM frameworks
LlamaIndex
RAG pipelines at batch pricing
The Journey: Request to Response
From loading dock to delivery — your cargo is in good hands.
Your App
POST /cargo/load
Queue Staging
Intelligent grouping
Batch (100)
Optimized delivery
Callback
Results delivered
Stop building batch infrastructure.
Start shipping AI features.
See what changes when you stop managing queues, retries, and batch windows yourself.
Without Convoy
You build and maintain your own batch processing pipeline. Every edge case is your problem.
→ 1 API call
With Convoy
Send a POST request and get results via callback. Convoy handles everything in between.
You could build this yourself.
But should you?
Every engineering team that builds batch processing in-house ends up maintaining it forever. Here's what you get out of the box with Convoy.
| Capability | DIY Batch Processing | Convoy |
|---|---|---|
| Time to first batch job | ❌ Weeks to months | ✅ Under a day |
| Infrastructure management | ❌ Queues, workers, scaling, monitoring | ✅ Fully managed — zero ops |
| Retry & error handling | ❌ Build from scratch | ✅ Built-in, automatic |
| Multimodal file handling | ❌ Build S3 plumbing + malware scanning yourself | ✅ Upload once, reference everywhere — malware-scanned |
| Cost optimization | ⚠️ Manual batching logic required | ✅ Intelligent auto-batching, ~40% savings |
| Scaling | ❌ Capacity planning & autoscaling config | ✅ Scales automatically with workload |
| Observability | ⚠️ Custom dashboards & alerting | ✅ Built-in tracking & status APIs |
| Minimum volume | ⚠️ Need volume to justify infra investment | ✅ No minimums — 1 request or 1 million |
Under the Hood
Built on battle-tested infrastructure for reliability at any scale.
REST API Gateway
A simple, well-documented API. Load cargo with a single POST request and receive a tracking ID instantly.
Intelligent Queue System
Requests are automatically grouped and optimized. No configuration needed — Convoy finds the best batch window.
Multi-Model AI Access
Access Claude, Llama, Mistral, Nova, and more through a single API. Switch models without changing your infrastructure.
Callback Delivery System
Results are delivered to your webhook as they complete. Real-time updates, zero polling required.
Security & Encryption
API key authentication, encrypted data in transit and at rest, and audit logging on every request.
Real-time Tracking
Monitor every batch job from submission to completion. Status APIs and dashboards give you full visibility.
Two ways to run Convoy.
Pick what fits your team.
A managed cloud platform for fast-moving teams, or a fully self-hosted enterprise deployment in your own AWS account.
Convoy Cloud
Managed SaaS — start in minutes
Sign up, get an API key, and start sending batch requests immediately. Convoy handles all infrastructure — queuing, batching, processing, and delivery.
Convoy Enterprise
Self-hosted in your own AWS account
Deploy Convoy into your own AWS account with a Terraform module or CloudFormation template. Full infrastructure — ECS Fargate, RDS, Cognito, ALB, Secrets Manager, CloudWatch observability, and the batch orchestration engine — production-ready in under a day.
All Aboard?
Ready to simplify your batch processing and start saving on AI costs? Get started in minutes.