Tessl finds Fable 5 marginally better than Opus 4.8
The evaluation of 917 agent skill tasks shows a 0.9 point overall score difference, with Fable 5 costing 73% more per task and refusing 26 tasks Opus completed.
See the latest news and media coverage for Tessl. We track all announcements, press releases, and industry mentions in real time, all in one place.
Agent enablement platform for AI-native software
tessl.ioLast updated
In short: Tessl advanced AI agent development through new evaluation tools and industry analysis, while navigating a complex AI infrastructure landscape.
The evaluation of 917 agent skill tasks shows a 0.9 point overall score difference, with Fable 5 costing 73% more per task and refusing 26 tasks Opus completed.
Evaluations across Claude Opus 4.7, Sonnet, Haiku, Codex, and Cursor show skills boost weaker models most, compressing performance gaps and optimizing costs. Haiku with skills excels for production.
The macOS and Windows app enables persistent AI agents for software tasks like debugging and code generation. It introduces visual management and Droid Computers.
Anthropic launches a managed service for deploying AI agents in production, handling long-running tasks, memory, and security. It supports companies like Notion and Asana. Pricing is consumption-based.
Tesla is recalling nearly a quarter of a million vehicles over a rear camera issue. The affected vehicles include certain 2017 Model 3, 2021-2023 Model...
FEATURE Tesla stock fell Friday, possibly on China disappointment, as it failed to make it three...
The notices of appeal were filed a month after Vice Chancellor Bonnie W. David dismissed the...
Tesla has never been a stock for the faint of heart. It trades on visions of the future, bet sizes that make traditional auto investors...
Track Tessl and your other target companies to get real-time alerts and weekly summaries delivered straight to your inbox.
Browse news for competitors to Tessl and other trending companies.