Understand this story
On November 1, 2026, Digital Life Kazik released an update to the AIHOT model leaderboard, increasing the publicly available evaluation sets it scrapes from 39 to 56, recalculating ELO rankings for all models using the original algorithm, and removing models that have not been officially launched, such as Gemini 4 Carbon. The updated leaderboard shows that Claude Opus 5.5 ranks first in the overall and coding categories, GPT-6 Astra ranks first in research and knowledge, and Gemini 4 Argon ranks first in office work. Kazik stated that this version of the rankings is fairly consistent with his understanding of each model. There are currently no further developments.
AI-generated from public reports
Latest development · Report
AIHOT model leaderboard expands to 56 evaluation sets and recalculates ELODevelopments
1 confirmed developments
- Source:English text is being prepared.
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 1
Kazike
AIHOT model leaderboard expands to 56 evaluation sets and recalculates ELO
The AIHOT model leaderboard has increased the number of public evaluation sets it collects from 39 to 56, recalculated ELO rankings for all models using the original algorithm, and removed models that have not been officially launched. Claude Opus 5.5 ranks first in the overall and coding categories, GPT-6 Astra ranks first in research and knowledge, and Gemini 4 Argon ranks first in office work.
Kazike
AIHOT Model Leaderboard Expands Evaluation Set to 56 and Recalculates ELO
The AIHOT model leaderboard has increased the number of publicly available evaluation sets it scrapes from 39 to 56, and recalculated ELO rankings for all models using the original algorithm, while removing models that have not been officially released (such as Gemini 4 Carbon). The author states that this version of the rankings aligns well with their understanding of each model.