Qwen 3 235B
A large-scale mixture-of-experts language model from Alibaba's Tongyi Qwen team, part of the Qwen3 series, with 235 billion total parameters and 22 billion activated parameters.
Timeline (last 90 days)
Qwen 3 235B and Gemma 4 26B both scored 0.6875 on resolve-plus, the lowest in the whole field.
Qwen 3 Coder 480B (0.875 vs 0.8125) clearly outperforms the larger general-purpose model Qwen 3 235B (0.8125 vs 0.6875), showing that models trained specifically for code have a real advantage on this type of task.
Qwen 3 235B achieved a perfect score of 112/112, causing the normal approximation interval to collapse to [1.00, 1.00]; switching to the Wilson interval yields a true range of approximately 0.97 to 1.00.
All Facts (3)
Qwen 3 235B and Gemma 4 26B both scored 0.6875 on resolve-plus, the lowest in the whole field.
50%UnverifiedQwen 3 Coder 480B (0.875 vs 0.8125) clearly outperforms the larger general-purpose model Qwen 3 235B (0.8125 vs 0.6875), showing that models trained specifically for code have a real advantage on this type of task.
50%UnverifiedQwen 3 235B achieved a perfect score of 112/112, causing the normal approximation interval to collapse to [1.00, 1.00]; switching to the Wilson interval yields a true range of approximately 0.97 to 1.00.
50%