Releasesby GlyphCog 0.21.0
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 .
Also shipped Jun 16, 2026Replicate in June 2026 →
QueryData adds support for parameterized secure views (PSVs) to help secure applications that use natural...
QueryData adds support for parameterized secure views (PSVs) to help secure applications that use natural language queries. For more information, see Secure and control access to application data . This feature is in Preview .
How to Use Claude Code with OpenRouter
Claude Code now supports routing through OpenRouter, providing budget controls and provider failover for Anthropic models.
How to Use OpenRouter With Any Coding Agent or AI Tool
Any tool using the OpenAI API can now integrate with OpenRouter by updating the base URL and API key.
Subagent: Let Your Model Delegate the Busywork
The new subagent tool allows frontier models to delegate self-contained tasks to smaller, faster worker models during generation.
Draft mode for V8.1 and new feature previews
Hi everyone - a few updates today! Draft mode for V8.1 Each draft mode generation creates 24 images at a lower resolution and quality. Click "Vary" on images you'd like to render at full-quality and full-resolution Draft jobs use half as many fast hours as V8.1 SD jobs (even though they generate 24 images!) Turn on draft mode by clicking the ⚡ button in the...
Redesigned Usage & Billing Page
The Usage & Billing page has been rebuilt to support usage attribution (the Settings "Overview" tab is now "Usage"). Filter by auth method, API key, and login username to attribute consumption to a specific credential or user. Normalized app names and revised charts improve readability and export.
More from CogEvery Cog release →
Cog 0.20.0
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.
Cog 0.19.3
Changelog 0780450 Bump version to 0.19.3 ( #2995 ) cac3e91 feat: experimental managed weights ( #2974 )
Cog 0.19.2
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 )
Cog 0.19.1
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.
Cog 0.22.0
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.
Cog 0.23.0
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.
Cog 0.23.1
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.