"xAI has introduced Grok 4 Fast, a new reasoning model designed for efficiency and lower cost. The model reduces average thinking tokens by 40% compared with Grok 4, which brings an estimated 98% decrease in cost for equivalent benchmark performance. It maintains a 2-million token context window and a unified architecture that supports both reasoning and non-reasoning use cases. The model also integrates tool-use capabilities such as web browsing and X search."
"In benchmark tests, Grok 4 Fast scores close to Grok 4 on GPQA, AIME, and HMMT, while outperforming Grok 3 Mini. On the LMArena Search Arena, its search variant ranked first with an Elo of 1163, and its text variant placed among the top in its category. Compared to similar models, Grok 4 Fast delivers higher efficiency than OpenAI's GPT-4 Turbo and Anthropic's Claude 3 Opus on cost-per-benchmark-point evaluations, while showing slightly lower raw accuracy on some high-end reasoning tasks."
xAI introduced Grok 4 Fast, a reasoning model optimized for efficiency and lower cost. The model reduces average thinking tokens by about 40% versus Grok 4, yielding an estimated 98% cost reduction for equivalent benchmark performance. It preserves a 2-million-token context window and uses a unified architecture for reasoning and non-reasoning tasks. Tool-use capabilities include web browsing and X search. Benchmark results show performance close to Grok 4 on GPQA, AIME, and HMMT and superiority over Grok 3 Mini. The search variant achieved first place on LMArena Search Arena with an Elo of 1163. Availability includes grok.com modes and xAI API access.
Read at InfoQ
Unable to calculate read time
Collection
[
|
...
]