AWS MLDeploy DeepSeek-R1 distilled models on Amazon SageMaker using a Large Model Inference container
Deploy DeepSeek-R1 distilled models on Amazon SageMaker using a Large Model Inference container
AWS MLBenchmarking Amazon Nova and GPT-4o models with FloTorch
Comparative benchmarking analysis of Amazon Nova and GPT-4o models using FloTorch for efficient AI model selection
Google Cloud5 ways ChromeOS uplevels the experience for Microsoft 365
Enhancing Microsoft 365 experience on ChromeOS with 5 key features and seamless integration
Google CloudScaNN for AlloyDB: The first PostgreSQL vector search index that works well from millions to billion of vectors
ScaNN for AlloyDB: Introducing a market-leading PostgreSQL vector search index for all use cases
Apple MLAn Efficient and Streaming Audio Visual Active Speaker Detection System
Developing real-time, efficient Active Speaker Detection system with reduced latency and memory usage
DatabricksAnnouncing Public Preview of AI/BI Genie Conversation APIs
Introducing the Public Preview of AI/BI Genie Conversation APIs to empower users with natural language data insights across various platforms.
DuolingoUsing Generative AI to Scale DuoRadio 10x Faster
Scaling DuoRadio 10x faster using generative AI for curriculum-driven script generation and automated audio production.
Duolingo6 inspiring stories from Duolingo English learners
Inspiring stories of Duolingo English learners overcoming challenges and achieving language goals.
OpenAIDetecting misbehavior in frontier reasoning models
Analyzing strategies for identifying errors in cutting-edge reasoning models
Ink and Switch04 · Opening the Pre-Alpha
Get early access to pre-alpha code for beelay-core, keyhive_core, and keyhive_wasm with caution.
DatabricksIntroducing New Governance Capabilities to Scale AI Agents with Confidence
Incorporating new governance capabilities to scale AI agents with confidence
SalesforceOptimizing AI Retrieval: How Agentforce Data Library Powers RAG with 99.99% Uptime
Improving AI retrieval performance with Agentforce Data Library's 99.99% uptime using RAG technology