[AINews] SpaceXAI launches Grok 4.5, first Opus-class model post Cursor acquisition
AINews/latent_space reports SpaceXAI launching Grok 4.5, its first Opus-class model since acquiring Cursor.
AI Storyline
3 items · 2 sources · 3 days
Operational story trace
Grok 4.5 launch
Follow in this browser to see new updates on your Live feed.
Current stateShipped · harness open-sourced
Latest change
One week after launch, SpaceXAI open-sourced the Rust agent harness, TUI, and tool layer behind its Grok coding CLI — the actual scaffolding behind the Jul 10 cost and hallucination numbers.
Earlier contextThe story so far
SpaceXAI shipped Grok 4.5 on Jul 9, its first Opus-class model since acquiring Cursor. A Jul 10 hands-on benchmark found the model cuts coding-agent cost 80% at near-frontier speed, though with a higher hallucination rate than rivals.
editor-curated · source-linked
Arc
Jul 9Jul 16 · now
LAUNCH · Jul 9SpaceXAI ships Grok 4.5, its first Opus-class model post Cursor acquisition1 source · show source ▾
BENCHMARK · Jul 10Hands-on benchmark: 80% cheaper coding-agent runs, near-frontier speed, more hallucinations1 source · show source ▾
Grok 4.5 Cuts Coding-Agent Cost 80%: Near-Frontier Speed, Higher Hallucinations - Tech Times
Tech Times publishes the first hands-on benchmark: 80% lower coding-agent cost at near-frontier speed, but more hallucinations than rivals.
OPEN SOURCE · Jul 16SpaceXAI open-sources the Rust harness, TUI, and tool layer behind its coding CLI1 source · scout · show source ▾
SpaceXAI Open-Sources Grok Build: The Rust Agent Harness, TUI, and Tool Layer Behind Its Coding CLI - MarkTechPost
MarkTechPost details the open-sourced Rust agent harness, TUI, and tool layer powering SpaceXAI's coding CLI — the infrastructure behind the Jul 10 benchmark, now inspectable directly.
Day 1 Thursday, Jul 9, 2026
[AINews] SpaceXAI launches Grok 4.5, first Opus-class model post Cursor acquisition
AINews/latent_space reports SpaceXAI launching Grok 4.5, its first Opus-class model since acquiring Cursor.
Day 2 Friday, Jul 10, 2026
Grok 4.5 Cuts Coding-Agent Cost 80%: Near-Frontier Speed, Higher Hallucinations - Tech Times
Tech Times publishes the first hands-on benchmark: 80% lower coding-agent cost at near-frontier speed, but more hallucinations than rivals.
Day 3 Thursday, Jul 16, 2026
SpaceXAI Open-Sources Grok Build: The Rust Agent Harness, TUI, and Tool Layer Behind Its Coding CLI - MarkTechPost
MarkTechPost details the open-sourced Rust agent harness, TUI, and tool layer powering SpaceXAI's coding CLI — the infrastructure behind the Jul 10 benchmark, now inspectable directly.
What to watch — open questions
- Does the 80% coding-agent cost cut hold up across benchmarks beyond Tech Times' initial write-up?
- How much does the higher hallucination rate cost in practice for agentic coding workflows — does it require extra verification steps that erode the price advantage?
- Does SpaceXAI publish its own benchmark numbers, or does independent testing stay the only source?
- What license governs the open-sourced harness, and does inspecting its tool layer explain why Grok 4.5 hallucinates more than rival coding agents?
How this thread was built
scout surfaced 1editor wrote the arc · 3 beatswatcher 1 status change
Storylines are threaded mechanically from the feed: stories that share a distinctive anchor across multiple days and sources. Each item links to its original source. The evidence trace, current state, and open questions are written by the editor routine and refreshed whenever a new beat lands.