Grok 4

Training compute

5×10²⁶ FLOP

Parameters

3T

Published

Jul 9, 2025

Grok 4 is an AI model developed by xAI (United States), first published in July 2025. It works in the language, multimodal and vision domain, on tasks such as language modeling/generation, question answering, search and 4 more. It counts among the frontier models: the systems trained with the most compute of their moment.

Training it took an estimated 5×10²⁶ FLOP of compute (estimation method: comparison with other models). The model has 3,000,000,000,000 parameters. The compute alone is estimated at $388M in 2023 dollars.

Access: API access. Its weights are not openly released. Epoch AI rates the confidence of this record as speculative.

Full record

Organization: xAI
Country of organization: United States
Domain: Language, Multimodal, Vision
Task: Language modeling/generation, Question answering, Search, Visual question answering, Character recognition (OCR), Image captioning, Quantitative reasoning
Training compute: 5×10²⁶ FLOP
Compute estimation method: Comparison with other models
Parameters: 3,000,000,000,000
Chips used: 200,000
Training cost (2023 USD): $388M
Model accessibility: API access
Open weights: No
Epoch confidence: Speculative

Benchmark results

01

Epoch Capabilities Index

ECI Score

147