AI BriefingOllama Limits Subscriptions and Issues Refunds Due to Caching Error
Key point
Ollama has acknowledged a usage limit overage issue caused by a model hosting caching error and is issuing refunds.
Details
Ollama recently confirmed an issue where some users' subscription sessions and weekly limits were being consumed faster than expected.
Upon investigation, a caching error during the model hosting process was found to be the cause, with this issue being particularly notable when using the DeepSeek-V4-Pro model.
Ollama stated that it will refund the past month's subscription fees to users affected by this error, and that it is taking technical measures to prevent the same issue from recurring in the future.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
NVIDIA Open-Sources Local AI Inference Router 'PAIR'
Redwood Defines 'NLS Depth' for AI Monitoring; Distinguishes Kimi-K3's High Depth from DeepSeek's OverestimationMagic AI Releases 'Magic V5 e24', Achieving Performance Similar to DeepSeek V4 Pro with 48x Fewer Computations