A user also claims Astra solved five Erdős problems, two under budget.
A user says GPT-6.1 Sol has saturated FrontierMath Tier 4, which they describe as exceptionally difficult even for research-level math in 2025. They call it 10 months of progress and also claim Astra solved five Erdős problems, two under budget.
318.7K
27 Sources, first seen ago
@haider1gpt-6.1 sol has saturated the FrontierMath Tier 4 benchmark that was considered exceptionally difficult even for research-level math just last year that's only 10 months of progress also Astra solved 5 erdős problems, and 2 of them under budget
@charliermarshRT @haider1: gpt-6.1 sol has saturated the FrontierMath Tier 4 benchmark that was considered exceptionally difficult even for research-leve…
@teortaxesTexOpenAI RL machine is fearsome they want math superintelligence and they're currently closer to it than anyone else This was the original idea of AGI, a couple years ago. Perhaps it's a more principled path to the goal than SWE agent-maxxing.@haider1gpt-6.1 sol has saturated the FrontierMath Tier 4 benchmark that was considered exceptionally difficult even for research-level math just last year that's only 10 months of progress also Astra solved 5 erdős problems, and 2 of them under budget@charliermarshRT @haider1: gpt-6.1 sol has saturated the FrontierMath Tier 4 benchmark that was considered exceptionally difficult even for research-leve…@teortaxesTexOpenAI RL machine is fearsome they want math superintelligence and they're currently closer to it than anyone else This was the original idea of AGI, a couple years ago. Perhaps it's a more principled path to the goal than SWE agent-maxxing.@haider1gpt-6.1 sol has saturated the FrontierMath Tier 4 benchmark that was considered exceptionally difficult even for research-level math just last year that's only 10 months of progress also Astra solved 5 erdős problems, and 2 of them under budget@charliermarshRT @haider1: gpt-6.1 sol has saturated the FrontierMath Tier 4 benchmark that was considered exceptionally difficult even for research-leve…@teortaxesTexOpenAI RL machine is fearsome they want math superintelligence and they're currently closer to it than anyone else This was the original idea of AGI, a couple years ago. Perhaps it's a more principled path to the goal than SWE agent-maxxing.@haider1gpt-6.1 sol has saturated the FrontierMath Tier 4 benchmark that was considered exceptionally difficult even for research-level math just last year that's only 10 months of progress also Astra solved 5 erdős problems, and 2 of them under budget@charliermarshRT @haider1: gpt-6.1 sol has saturated the FrontierMath Tier 4 benchmark that was considered exceptionally difficult even for research-leve…@teortaxesTexOpenAI RL machine is fearsome they want math superintelligence and they're currently closer to it than anyone else This was the original idea of AGI, a couple years ago. Perhaps it's a more principled path to the goal than SWE agent-maxxing.—
Not ranked yet
—
Not ranked yet
Comments