金融科技独角兽 Revolut 转向智谱 GLM，编程账单约降至十分之一
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.5 Articles
5 Articles
Revolut claims to have achieved a 90% reduction in AI costs, from the commercial services of OpenAI and Anthropic to Chinese GLM models on dedicated infrastructure. David Abitbol, head of AI operations, explains it. Revolut key points uses GLM 5.3 out of 56 NVIDIA GPUs B200. The processing costs on average $0.30 per million [...]
The European online bank runs GLM, an open model designed in China, on dedicated servers. It claims to have divided its AI bill by ten while delivering faster. The conference was held in mid-October in New York, but its echo only came with its re-publication on YouTube on 29 October.
金融科技独角兽 Revolut 转向智谱 GLM，编程账单约降至十分之一
ChainCatcher 消息，欧洲估值最高的金融科技独角兽 Revolut 已将人工智能公司智谱的 GLM 设为约 2000 名工程师的默认 AI 编程模型，Claude Code、Codex 等工具退居备选，主要留给复杂任务。Revolut AI 运营负责人 David Abitbol 透露，公司此前给 Claude Code 设置每人每月 300 美元的额度，但对整体效率的帮助有限。今年 4 月，公司开始自行部署 GLM，并向工程师开放不限量 Token。目前，Revolut 使用约 56 张英伟达 B200 运行 GLM 5.3，折算每百万 Token 的综合成本约 0.30 美元，相关账单仅为此前使用 Anthropic、OpenAI 的十分之一。根据公司内部统计，工程师的编码用时指标改善约...
Coverage Details
Bias Distribution
- There is no tracked Bias information for the sources covering this story.
Factuality
To view factuality data please Upgrade to Premium
