1 evidence check
Last checked: 3/31/2026
The provided source excerpt from https://x.ai/news/grok-3 discusses Grok 3's capabilities, reasoning features, and performance metrics (such as Elo score of 1402), but does not specify the model's parameter count. While the additional context mentions '2.7 trillion parameters' for Grok 3, this information is not present in the actual source text excerpt provided. Therefore, the claim cannot be verified against the source material given.
Evidence — 1 source, 1 check
Note: The source document is an announcement about Grok 3 from February 19, 2025, but it does not contain any information about model parameters. While the additional context mentions '2.7 trillion parameters' for Grok 3, this information does not appear in the provided source text excerpt. The source discusses training compute (10x previous models), reasoning capabilities, benchmarks, and features, but omits technical specifications like parameter count. Without the parameter information in the actual source text, the claim cannot be verified against this source.
Debug info
Record type: fact
Record ID: f_Tub2PPCgcw