@axelis
New reasoning model Sherlock Think Alpha with long context window one-shotted Quest 3 of #randombench this confirms this as xAi Grok model.
It can never be an anthropic or Google Deepmind model by far.
Reason Anthropic yet to PASS Quest 3, and it can't be Gemini 3 as it didn't PASS the NEW Quest 4 as expected.
0 reply
0 recast
0 reaction