On Thursday, Anthropic released Opus 4.8, the most recent model of its most superior publicly accessible mannequin. The mannequin is on the market all over the place, with customary pricing on the identical stage because the earlier Opus launch.
The brand new mannequin comes simply 41 days after Opus 4.7 was launched, a a lot sooner improve cycle than regular for Anthropic. (The newest Sonnet and Haiku fashions are three and 7 months outdated, respectively.) The quick turnaround could have one thing to do with the chilly reception to Opus 4.7, which some customers found disappointing.
That interval has additionally seen vital new releases for OpenAI’s Codex and Google’s Gemini Flash model, rising the stress on Anthropic to maintain tempo.
Opus 4.8 comes with the anticipated best-in-class benchmark outcomes, however there’s additionally specific consideration to how the mannequin manages dangerous or unsure knowledge. Within the launch submit, Anthropic’s early testers discovered that the brand new mannequin is “extra more likely to flag uncertainties about its work and fewer more likely to make unsupported claims.”
Echoing this level, a testimonial from Bridgewater associates stated the most important distinction within the improve was “Opus 4.8’s tendency to proactively flag points with the inputs and outputs of an evaluation, one thing different fashions routinely missed and left to the customers to catch.”
Along with the brand new mannequin, Anthropic launched a function known as Dynamic Workflows, which might be accessible in analysis preview. The system is designed to assist bigger fashions like Opus handle advanced duties throughout lots of of parallel subagents.
“Claude Code alongside Opus 4.8 can now perform codebase-scale migrations throughout lots of of hundreds of traces of code from kickoff to merge, with the present check suite as its bar,” the submit explains.
Anthropic remains to be holding again its most superior Mythos mannequin after a tentative preview last month raised cybersecurity considerations. Nonetheless, the corporate hinted in right now’s Opus launch that the Mythos preview interval would possibly quickly finish, as soon as essential safeguards are full.
“We’re making swift progress on creating these safeguards and count on to have the ability to carry Mythos-class fashions to all our prospects within the coming weeks,” the corporate wrote.
Whenever you buy by way of hyperlinks in our articles, we may earn a small commission. This doesn’t have an effect on our editorial independence.