xAI’s Grok 4.5 Challenges AI Model Market with Aggressive Pricing

xAI has launched Grok 4.5, a new AI model trained on tens of thousands of Nvidia GB300 GPUs, aiming to challenge the established market leaders like OpenAI’s GPT-5.5 and Anthropic’s Fable 5, particularly in coding and agentic tasks. While benchmark results present a mixed picture regarding its raw performance, Grok 4.5’s aggressive pricing strategy positions it as a significant contender, especially for developers and businesses in cost-sensitive markets like India.

Initial benchmarks indicate that Grok 4.5 performs competitively in certain areas. On Terminal Bench 2.1, which assesses complex command-line tasks, Grok 4.5 achieved 83.3%, nearly matching GPT-5.5 (83.4%) and closely trailing Fable 5 (84.3%). However, its performance diverges in more demanding software engineering evaluations. On DeepSWE 1.1, designed to resolve real GitHub issues, Grok 4.5 scored 53%, notably behind GPT-5.5’s 67% and Fable 5’s 70%. For SWE Bench Pro, a collection of tougher software engineering problems, it reached 64.7%, surpassing Opus 4.8 in some configurations but falling short of Fable 5’s 80.4%.

Training and Efficiency Focus

xAI emphasized its rigorous approach to data quality during Grok 4.5’s development, employing heavy data filtering, deduplication, and domain-specific selection. The model’s reinforcement learning phase involved hundreds of thousands of tasks, primarily focused on software engineering, with automated scoring. The training infrastructure was built for asynchronous learning, allowing agentic runs to span many hours while training continued in parallel, optimizing resource utilization.

Key facts:

Feature	Detail
Model	Grok 4.5
Developer	xAI
Input Token Cost	$2 per million
Output Token Cost	$6 per million
Availability	Now via Grok Build, Cursor, xAI console; EU launch mid-July

Disruptive Pricing Model

The most compelling aspect of Grok 4.5 is its pricing. At $2 per million input tokens and $6 per million output tokens, it significantly undercuts its rivals. For comparison, Opus 4.8 costs $5 for input and $25 for output per million tokens, while Fable 5 charges $10 for input and $50 for output. GPT-5.5 and GPT-5.6 are priced at $5 for input and $30 for output. xAI also claims Grok 4.5 uses 4.2 times fewer tokens than Opus 4.8 on SWE Bench Pro tasks and delivers results at 80 tokens per second. This combination of lower per-token pricing and higher token efficiency could make Grok 4.5 a highly attractive option for developers and businesses looking to optimize their AI operational costs.

Implications for the Indian Market

This aggressive pricing strategy, mirroring that of some Chinese AI vendors, could have a substantial impact on the adoption of advanced AI models in India. Indian startups and tech companies, often operating with tighter budgets, could find Grok 4.5’s cost-effectiveness a decisive factor, enabling them to deploy sophisticated AI capabilities without prohibitive expenses. This could accelerate AI innovation and application development across various sectors in India, from software development to automated customer service.

Grok 4.5 is currently available through Grok Build, Cursor, and the xAI console, with plugins for popular applications like Word, PowerPoint, and Excel. While it is not yet available in the EU, xAI targets a mid-July launch.

Source: The Decoder (https://the-decoder.com/grok-4-5-is-so-cheap-compared-to-fable-5-and-gpt-5-5-that-benchmark-gaps-may-not-matter-much/)