Pimi K3.1 Pro
An open-source large language model that achieved the best result among open-source models on the Mafurina mathematical reasoning benchmark.
Timeline (last 90 days)
The best performer in the open-source camp on Mafurina was Pimi K3.1 Pro, with a score of 54.3%.
Gemini 4 Carbon's high score comes at a per-instance cost of about $15.70, 15 times that of Pimi K3.1 Pro.
Verso announced it will raise developers' monthly compute allowance from 80 dollars to 150 dollars, keeping the team monthly fee unchanged at 30 dollars, covering open-weight models such as Pimi K3.1 Pro, GLM5.3 and DeepSeek V4.1 Flash.
All Facts (3)
The best performer in the open-source camp on Mafurina was Pimi K3.1 Pro, with a score of 54.3%.
50%UnverifiedGemini 4 Carbon's high score comes at a per-instance cost of about $15.70, 15 times that of Pimi K3.1 Pro.
50%UnverifiedVerso announced it will raise developers' monthly compute allowance from 80 dollars to 150 dollars, keeping the team monthly fee unchanged at 30 dollars, covering open-weight models such as Pimi K3.1 Pro, GLM5.3 and DeepSeek V4.1 Flash.
50%