Grok 4.5
Grok 4.5 is a coding-focused large language model released by xAI in July with 1.5 trillion parameters, positioned as a competitive alternative to Claude Opus at significantly lower cost and token usage. Recent benchmarks show it matches Claude's coding performance while using roughly 4.2 times fewer tokens at less than half the price, and it has demonstrated strong results in agent-based tasks and coding benchmarks like SWE-bench (~70%) and the Basecamp project builds. SpaceX AI has leveraged Grok 4.5 as the backbone for Grok Build, an open-sourced terminal coding agent, and the model operates at $2/$6 per million input/output tokens with throughput of 80 tokens per second.
Updated 3 August 2026
Specifications
No specifications recorded yet.
Latest developments
Are AI labs pelicanmaxxing?
Simon Willison · 1 week ago ·
44
LWiAI Podcast #252 - GPT 5.6, Grok 4.5, Nemotron-Labs-Diffusion, AI 2040
Last Week in AI · 1 week ago ·
43
Agent swarms and the new model economics
TLDR · 1 week ago ·
33
I trust Claude for everything. This test made me rethink that.
The New Stack · 2 weeks ago ·
5
Musk open-sourced Grok Build to fight Anthropic. Anthropic pays him $1.25 billion a month.
The New Stack · 2 weeks ago ·
33
SpaceXAI Open-Sources Grok Build: The Rust Agent Harness, TUI, and Tool Layer Behind Its Coding CLI
MarkTechPost · 2 weeks ago ·
48
2026
Study finds no evidence of AI labs optimizing models for pelican-on-bicycle image generation Research publication
Multiple AI model releases and regulatory developments across OpenAI, Anthropic, and competitors Other
xAI open-sources Grok Build coding agent and releases Grok 4.3 on Amazon Bedrock Open source release
Claude Sonnet 5 Release and Tokenizer Efficiency Comparison with Competitors Model release
xAI launches Grok 4.5, an Opus-class large language model developed in partnership with Cursor Model release
OpenAI releases GPT-5.6 in multiple variants alongside frontier AI model releases from competitors Model release
- Are AI labs pelicanmaxxing?
- "Drawing" the Mona Lisa with GPT-5.6, Claude, Gemini, and Grok
- LWiAI Podcast #252 - GPT 5.6, Grok 4.5, Nemotron-Labs-Diffusion, AI 2040
- Agent swarms and the new model economics
- I trust Claude for everything. This test made me rethink that.
- Musk open-sourced Grok Build to fight Anthropic. Anthropic pays him $1.25 billion a month.
- SpaceXAI Open-Sources Grok Build: The Rust Agent Harness, TUI, and Tool Layer Behind Its Coding CLI
- Basecamp Bench
- The Sequence Radar #893: Last Week in AI: GPT-5.6, Grok 4.5, Muse Spark 1.1 and the Post-Chatbot Stack
- Deep Learning Weekly: Issue 463
- AI #176 Part 1: Doing It Live
- Grok x Cursor
- Introducing Grok 4.5
- SpaceXAI releases Grok 4.5, which Elon describes as an 'Opus-class model'
- [AINews] SpaceXAI launches Grok 4.5, first Opus-class model post Cursor acquisition
Relationships
Products & technology
- xAI develops this model · 4 sources
- SpaceX develops this model · 3 sources
- SpaceXAI develops this model · 2 sources
- Cursor deploys this model · 1 source
- Grok Build derived from this model · 1 source