Grok Build 0.2.92 adds Grok 4.5

// 93d agoMODEL RELEASE

Grok Build 0.2.92 adds Grok 4.5

The new Grok 4.5 AI model has been integrated into the Grok Build terminal development tool. Users must update to release 0.2.92 and have SuperGrok Heavy tier access to participate in the early access program.

// ANALYSIS

Gating the model behind a "SuperGrok Heavy" tier suggests a premium pricing structure or resource constraints for running the model at scale.

* Grok Build 0.2.92 is required to enable Grok 4.5 integration.

* Access is limited to users with SuperGrok Heavy tier privileges.

* This marks a swift integration of Grok 4.5 into developer-focused agentic coding toolchains.

// TAGS

grok-4.5grok-buildxaispacexaillmcoding-assistantagentai-codingcode-generation

DISCOVERED

93d ago

2026-07-08

PUBLISHED

93d ago

2026-07-08

RELEVANCE

8/ 10

AUTHOR

mark_k

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

INFRA1h ago

boat reports 7.7x ARR growth after YC

YC-backed boat says it grew ARR 7.7x, DAUs 6x, daily VM time 8.6x, and weekly new trials 5.6x in two months, while adding 22 percentage points of NRR with the same four-person team. Its product provides persistent Ubuntu VMs for long-running AI agents, with Docker, Chrome, desktop access, snapshots, and APIs.

TUTORIAL1h ago

Claude Code Clarifies MCP Server Scopes

Anthropic’s guide explains how Claude Code stores MCP servers across local, user, and project scopes, including how `.mcp.json` enables team sharing. Project-scoped servers require approval before they can run, protecting developers from untrusted repository configuration. [Official documentation](https://code.claude.com/docs/en/mcp-quickstart)

BENCHMARK2h ago

Membase Reports 93% Agent-Memory Scores

Membase is an evidence-centered memory system for AI agents, organizing episodic, semantic, and procedural memory. It reports scores of 93.12% on LoCoMo, 92.60% on LongMemEval-S, and 92.20% on DMR.