Also onPaydays(Pay schedule)
Bug fixes - Predictions now fail when an output upload fails. Previously, a file upload that timed out or was rejected could leave a prediction marked succeeded with output: null, or silently drop files from list outputs. Any failed upload now fails the prediction with an error, and cancellation still reports canceled.
New features GPU compatibility warnings in cog doctor . cog doctor now warns when the resolved PyTorch and CUDA wheel ships no kernels for the local GPU, catching images that build successfully but fail when they run CUDA operations. ( #3104 ) OpenTelemetry tracing.
Breaking changes Local servers now bind to 127.0.0.1 instead of 0.0.0.0 . cog serve , cog run -p , cog predict , and cog train no longer expose the prediction endpoint to your whole network by default. To get the old behavior back, use cog serve --host 0.0.0.0 or cog run -p 0.0.0.0:8888 , -p now accepts host:port . ( #2812 ) New features Browser playground.
New features Server-Sent Event prediction streams. HTTP prediction requests can now ask for Accept: text/event-stream to receive start , output , log , metric , and terminal completed events for predictors that explicitly opt in with @streaming / @cog.streaming .
New features cog run command. The cog predict command has been renamed to cog run with full backward compatibility. cog predict still works as an alias. ( #3015 ) Model refs for cog push and weights commands. You can now reference models by name (e.g., r8.im/user/model ) instead of full image URLs when pushing or managing weights.
Changelog 0780450 Bump version to 0.19.3 ( #2995 ) cac3e91 feat: experimental managed weights ( #2974 )
Changelog 427d486 Bump version to 0.19.2 ( #2984 ) d9c2b14 fix: cap fuzz test input size to prevent timeout ( #2981 ) 4269575 fix: support typing_extensions TypedDict at runtime ( #2983 )
Bug fixes Support for TypedDict in schema generation. Fixed an issue where TypedDict type annotations would cause schema generation to fail. ( #2978 ) Build order fix for resource exhaustion. Reordered coglet wheel build to run after cog/sdk builds to prevent resource exhaustion during release builds. ( #2977 ) Maintenance Removed dead Go code.
Agent skills that give coding assistants expert knowledge about finding, comparing, running, and prompting AI models on Replicate.
Nano Banana Pro can now fall back to another image generation model when Google's API is at capacity.
Replicate's MCP server can now be discovered via the official MCP Registry.
Filter the list of predictions API to show only predictions created through the web interface.
A collection of updates to Replicate for the week ending December 19, 2025.
A collection of updates to Replicate for the week ending December 5, 2025.
A collection of updates to Replicate for the week ending November 21, 2025.
An experimental feature that lets language models write and execute TypeScript code in a sandboxed environment.
A collection of updates to Replicate for the week ending November 7, 2025.
Enterprise customers can now get notified when deployments fail to start and customize how long to wait before timing out.
A collection of updates to Replicate for the week ending October 24, 2025.
Set predictions to automatically cancel if they don't finish within a specified duration.
A collection of updates to Replicate for the week ending October 10, 2025.
View and download invoices directly from your billing settings page
You can now update model properties using the API.
A collection of updates to Replicate for the week ending September 26, 2025.
| Release-notes page | Site | Feed |
|---|---|---|
| Changelog replicate.com | replicate.com | None |
| Introducing a new Cog runtime replicate.com | replicate.com | None |
| Run all models with the same API endpoint replicate.com | replicate.com | None |
| Stable diffusion has release notes replicate.com | replicate.com | None |
Releasesby Glyph