vLLM intègre le support des TPU et améliore les performances de service
vllm integrates TPU support and boosts model serving performance
3 comptes indépendants
20 messages
6 articles
3 623 interactions
vllm
@GithubProjects
@dani_avila7
@rohanpaul_ai
@vllm_project
@NetEaseYouDaoAI
@aiDotEngineer
@fraserpricee
@inferact
netease-youdao/Confucius4-R2T2 · Hugging Face huggingface.co
netease-youdao/Confucius4-R2T2 huggingface.co