Perplexity Research research.perplexity.ai

https://research.perplexity.ai/articles/hosting-qwen-on-blackwell

research.perplexity.ai·3d

https://research.perplexity.ai/articles/cutedsl-at-perplexity

research.perplexity.ai·1w

https://research.perplexity.ai/articles/designing-refining-and-maintaining-agent-skills-at-perplexity

research.perplexity.ai·1w·Hacker News

https://research.perplexity.ai/articles/advancing-search-augmented-language-models

research.perplexity.ai·3w·Hacker News

https://research.perplexity.ai/articles

research.perplexity.ai·3w

RL Training For Math Reasoning

research.perplexity.ai·10w

Accelerating Sonar Through Speculation

research.perplexity.ai·10w

Lower Latency and Higher Throughput with Multi-node DeepSeek Deployment

research.perplexity.ai·10w

Disaggregated Prefill and Decode

research.perplexity.ai·10w

Efficient and Portable Mixture-of-Experts Communication

research.perplexity.ai·10w

High-Performance GPU Memory Transfer on AWS Sagemaker Hyperpod

research.perplexity.ai·10w

BrowseSafe: Understanding and Preventing Prompt Injection Within AI Browser Agents

research.perplexity.ai·10w

Architecting and Evaluating an AI-First Search API

research.perplexity.ai·10w

RDMA Point-to-Point Communication for LLM Systems

research.perplexity.ai·10w

GPT-OSS on Day 0

research.perplexity.ai·10w

pplx-embed: State-of-the-Art Embedding Models for Web-Scale Retrieval

research.perplexity.ai·11w·Hacker News

Evaluating Deep Research Performance in the Wild with the DRACO Benchmark

research.perplexity.ai·14w

Weight Transfer for RL Post-Training in under 2 seconds

research.perplexity.ai·16w·Hacker News

Enabling Trillion-Parameter Models on AWS EFA

research.perplexity.ai·27w·Hacker News

Log in to enable infinite scrolling