Grok 4.5 launches with competitive pricing but mixed benchmark results against GPT-5.5 and Fable 5
xAI has released Grok 4.5, a new AI model trained on tens of thousands of Nvidia GB300 GPUs, targeting coding, agentic tasks, and knowledge work. Benchmark results show a mixed picture: Grok 4.5 nearly matches GPT-5.5 and Fable 5 on Terminal Bench 2.1 (83.3% vs 83.4% and 84.3%), but trails significantly on DeepSWE 1.1 (53% vs 67% for GPT-5.5 and 70% for Fable 5). However, Grok 4.5 is dramatically cheaper at $2 per million input tokens — a fraction of competitors — and uses 4.2 times fewer tokens than Opus 4.8. The article argues that the cost advantage may outweigh benchmark performance gaps for many use cases. EU availability is expected in mid-July.
Key quotes
Grok 4.5 scores 83.3%, nearly matching GPT 5.5 (83.4%) and trailing Anthropic's Fable 5 (84.3%) by just one point.
On DeepSWE 1.1, which measures the ability to resolve real GitHub issues, Grok 4.5 hits 53%, well behind OpenAI's GPT-5.5 at 67% and Fable 5 at 70%.
At $2 per million input tokens, it costs a fraction of the competition.
The model was trained on tens of thousands of Nvidia GB300 GPUs and targets coding, agentic tasks, and knowledge work.
You might also wanna read
Grok 4.5 vs GPT-5.5 vs Claude Opus 4.8 vs Fable 5: A controlled coding benchmark
A comparison of four leading AI coding models — Grok 4.5, GPT-5.5, Claude Opus 4.8, and Claude Fable 5 — tasked with building the same three
tryai.dev·9h ago
Grok Build 0.1 0616: Comprehensive AI Model Benchmarking and Competitive Analysis
This article provides an in-depth analysis of xAI's Grok Build 0.1 0616, comparing it against other AI models across key metrics including i
artificialanalysis.ai·15d ago
xAI Launches Grok 4 Fast AI Model with Free Access and Competitive Pricing
xAI has launched Grok 4 Fast, a new AI model that offers high speed (190 tokens per second), low cost ($0.20/1M input tokens, $0.50/1M outpu
Product Hunt·4mo ago
xAI Launches Grok 4 Fast: Cost-Efficient Reasoning Model with 2M Token Context
xAI has launched Grok 4 Fast, a new cost-efficient reasoning model that builds on learnings from Grok 4. The model delivers frontier-level p
x.ai·9mo ago
Grok 4.5 Raises the Bar for Frontier AI With High-Speed Reasoning, Token Efficiency, and Real-World Enterprise Automation
SpaceXAI launches Grok 4.5, its flagship AI model for coding and engineering tasks
SpaceXAI has launched Grok 4.5, their most advanced AI model designed for coding, agentic tasks, and knowledge work. Trained alongside Curso
x.ai·14h agoSpaceXAI launches Grok 4.5, its flagship AI model for coding and engineering tasks
SpaceXAI has launched Grok 4.5, their most advanced AI model designed for coding, agentic tasks, and knowledge work. Trained alongside Curso
x.ai·14h agoSpaceXAI launches Grok 4.5, its flagship AI model for coding and engineering tasks
SpaceXAI has launched Grok 4.5, their most advanced AI model designed for coding, agentic tasks, and knowledge work. Trained alongside Curso
x.ai·14h agoSpaceXAI launches Grok 4.5, its flagship AI model for coding and engineering tasks
SpaceXAI has launched Grok 4.5, their most advanced AI model designed for coding, agentic tasks, and knowledge work. Trained alongside Curso
x.ai·14h ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.