OpenCode 1.18.32 adds Grok 4.7, TypeSafe Jev

// 2h agoPRODUCT UPDATE

OpenCode 1.18.32 adds Grok 4.7, TypeSafe Jev

ANNOUNCEMENT PRODUCT GITHUB PRODUCT HUNT X

OpenCode released version 1.18.32, updating its Zen, Bedrock, Core, and Data sync components. The release adds Grok 4.7 and TypeSafe AI Jev support, hoists Bedrock tool images for non-multimodal models, and resolves Node package entrypoint issues.

// ANALYSIS

While an incremental patch, v1.18.32 delivers critical compatibility shims and stability fixes that prevent silent failures across heterogeneous model backends.

–Incorporating Grok 4.7 and TypeSafe AI Jev structured outputs keeps pace with rapidly evolving model options and structured generation workflows.
–Image hoisting for non-multimodal models on Bedrock smartly bridges capability mismatches during tool-use execution.
–Fixing Node entrypoint resolution and adding paginated retries addresses core developer friction and strengthens sync resilience.

// TAGS

opencodedevtoolsai-codinggrokaws-bedrockopensource

DISCOVERED

2h ago

2026-09-23

PUBLISHED

2h ago

2026-09-23

RELEVANCE

6/ 10

AUTHOR

OpenCodeLog

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

UPDATE1h ago

CookMyMeme adds MCP for autonomous token launches

Onchain token launcher CookMyMeme announced the rollout of agentic AI capabilities, exposing its core suite of token creation, market monitoring, and trading tools to autonomous agents through a dedicated API and Model Context Protocol (MCP) server. The integration allows AI agents and humans to share the same onchain execution rails across networks such as Solana and Base, automating workflows that range from discovering tokens and analyzing markets to drafting launches, generating branding assets, and constructing trades.

UPDATE1h ago

OpenCode 2.0.4 ships with MCP and Cloudflare updates

OpenCode version 2.0.4 upgrades its MCP client to SDK v2 and adds support for the 2026-07-28 protocol. The release also preserves prompt caching when changing reasoning effort mid-session, expands Cloudflare AI Gateway API support, and cleans up the v2 HTTP API.

UPDATE1h ago

classifier.dev slashes inference costs with fraud protection

The creator of classifier.dev rolled out automated fraud protection across the platform, reducing API hosting expenses from $15/hour to under $1/hour. The safeguards filter abusive automated traffic and bot scraping, preserving free anonymous access to keyless zero-shot classification endpoints without mandatory registration.