genuinely like 90% of online discourse right now
GPT-6.1-Sol gets #2# on eyebench-v3, stealing Opus-5.5s short-lived spot!
It's near-astra performance for ~3.8x cheaper, and ~8x cheaper than Opus-5.5 while scoring better.
Astra-max scores 95% on eyebench-v3.
At half the cost (outputting ~3.8x less tokens) of Sol-max, it scores nearly double.
this is astra btw.
first model to visibly leave me at a loss for words
Fable-5.1-max gets #4# on eyebench-v3!
This is the biggest jump I've seen from anthropic on this bench.
It's beaten only marginally by 5.6-terra-max and handidly by 5.6-sol-max.
Show more
claude fable 5 (left) vs gpt-5.5 (right)!
"makes a twitter clone in minecraft"
holy anthropic mog
