Blog
Product updates, engineering deep-dives, and practical guides on AI batch processing.
Resume Your Durable Lambda Directly: No Webhook Required
Convoy can now resume your AWS Lambda durable function directly — a scoped IAM role replaces the webhook endpoint, receiver Lambda, and signing scheme.
Read postEight AWS-Native Delivery Targets for Your Batch Results
Convoy now delivers batch results straight into SQS, Kinesis, EventBridge, S3, Lambda, and Step Functions — your IAM role, no webhook endpoint required.
Batch Multimodal Inference: Images and PDFs at Up to 50% Off
Upload images and PDFs once, reference them by ID in batch requests, and process vision and document workloads at up to 50 percent off real-time prices.
Your AI Bill Is Too High, and It's One Setting
Most AI workloads don't need real-time responses, yet teams pay real-time prices. One setting can cut costs up to 50 percent on supported workloads.
Convoy Is Live: Stop Building Batch Infrastructure
Convoy is live. Send AI requests to one API, we batch them automatically, and cut inference costs up to 50 percent with no batch infrastructure to manage.
When to Use Batch Inference (and When Not To)
A practical framework for deciding which AI workloads belong on batch inference APIs, with real cost math, a migration checklist, and pitfalls to avoid.