Understand this story
On November 21, ARC Prize testing showed that Gemini 4 Hydrogen scored 92.5% on ARC-AGI-2 and 98.5% on ARC-AGI-1, at a cost per task of $1.26 and $0.18, respectively. At the high reasoning setting, its performance matched or exceeded Gemini 3.8 Flash while using approximately 52% fewer reasoning tokens. ARC-AGI-3 testing could not be completed due to multi-action function calling and API issues such as MALFORMED_FUNCTION_CALL. There are currently no other subsequent developments supported by factual evidence.
AI-generated from public reports
Latest development · Report
ARC Prize Tests Gemini 4 Hydrogen: Scores 92.5% on ARC-AGI-2Developments
1 confirmed developments
- Source:ARC Prize
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 21
ARC PrizeFirst-partySelected
ARC Prize Tests Gemini 4 Hydrogen: Scores 92.5% on ARC-AGI-2
ARC Prize found that Gemini 4 Hydrogen scored 92.5% on ARC-AGI-2 and 98.5% on ARC-AGI-1, at a cost per task of $1.26 and $0.18, respectively. At the high reasoning level, it matched or exceeded Gemini 3.8 Flash while using about 52% fewer reasoning tokens; ARC-AGI-3 testing could not be completed because of API issues including multi-action function calls and MALFORMED_FUNCTION_CALL errors.
AI Trends