Changelog

See our latest feature releases, product improvements and bug fixes

Jun 30, 2026

Connect coding agents to Baseten

Connect your coding agent to the Baseten MCP server and install the Baseten skill to manage your workspace from your agent. Your agent can deploy and promote models, tune autoscaling, pull logs, and...

Jun 29, 2026

Configure scale-down rate

You can now cap how aggressively the autoscaler removes replicas when traffic drops. Set max_scale_down_rate between 1% and 50% (default 50%) to limit the share of excess replicas removed at each...

Jun 25, 2026

Log downloads

Download all of a deployment's logs for a chosen time range and filters straight from the dashboard. Baseten runs the export as a background job and returns a CSV or JSON file covering up to 7 days...

Jun 24, 2026

Model Deprecation (DeepSeek v3.1, MiniMax m2.5)

The DeepSeek v3.1 and MiniMax M2.5 Model API(s) were deprecated at 5pm PT on June 24th.

Jun 22, 2026

Filter and stream model logs from the CLI

truss model-logs now takes filter flags so you can pull exactly the logs you need. Scope the window with --since, --start, and --end, filter by --min-level or message text with --includes and...

Jun 16, 2026

GLM 5.2 available on Baseten

You can start sending requests to GLM 5.2 today through our Model APIs by calling the OpenAI-compatible endpoint with your Baseten API key. For larger workloads, dedicated deployments are available.

Jun 16, 2026

Kimi K2.7 Code available on Baseten

You can start sending requests to Kimi-K2.7-Code today through our Model APIs by calling the OpenAI-compatible endpoint with your Baseten API key. For larger workloads, dedicated deployments are...

Jun 12, 2026

New sidebar navigation

We just rolled out a new sidebar nav across Models, Chains, Model APIs, and Training making it easier to navigate between resources, access environments and deployments, and quickly jump to actions...

Jun 11, 2026

Container restart tracking

Each deployment's metrics dashboard now includes a graph showing model container restarts over time. Spikes point to crashes in your model code, out-of-memory kills, or failed health checks.

Jun 10, 2026

vLLM and SGLang metrics

Baseten now surfaces engine-native metrics for models served with vLLM or SGLang directly in the Metrics tab.

Page 3 | Changelog