Goose 1.37.0 drops mid-session model switching

// 2h agoPRODUCT UPDATE

Goose 1.37.0 drops mid-session model switching

Goose v1.37.0 has launched, focusing on enhancing developer workflows across terminals, desktops, sessions, and agent pipelines. The update introduces the ability to switch models mid-session in the CLI using the `/model` command, as well as a `/goal` command that allows the agent to self-evaluate its progress and performance on tasks.

// ANALYSIS

Mid-session model switching is a crucial addition that addresses the balance between cost and capability, allowing developers to dynamically alternate between models depending on the complexity of the subtask.

* Swap models seamlessly mid-session using `/model` to match LLM capacity to task difficulty.

* The `/goal` command introduces self-evaluation, which could significantly increase autonomous task completion rates by reducing reliance on manual verification.

* The focus on multi-modal, CLI, terminal, and desktop integration positions Goose as a highly versatile local-first assistant compared to web-locked IDE wrappers.

// TAGS

gooseagentdeveloper-toolsblockopen-source

DISCOVERED

2h ago

2026-06-03

PUBLISHED

2h ago

2026-06-03

RELEVANCE

8/ 10

AUTHOR

goose_oss

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

UPDATE1h ago

Factory Router cuts agent inference costs

Factory has launched Factory Router, an intelligent model-routing feature that dynamically dispatches developer queries based on task complexity. The system directs routine tasks to smaller models while reserving frontier LLMs for complex modifications, reducing token costs by 20% to 25%.

UPDATE1h ago

Grok Imagine Video hits Vercel AI Gateway

Grok Imagine Video, an image-to-video generation model, has been integrated into the Vercel AI Gateway. It is being touted as the top-performing image-to-video model currently available on the arena.ai leaderboard.

UPDATE1h ago

Vercel AI Gateway adds Grok Video 1.5

Vercel has added support for xAI's Grok Imagine Video 1.5 preview to the Vercel AI Gateway and AI SDK. Built on xAI's Aurora engine, the model generates synchronized audio—including speech, sound effects, and music—alongside video frames in a single inference pass for clips up to 15 seconds long.