Sublime

How to understand Github code repos with LLMs in 2 mins. Even after 10yrs of engineering, dissecting a large codebase is daunting. 1 Dump the code into one big file 2 Feed it to Gemini-1.5Pro (2M context) 3 Ask it anything. Here's, I dissect DeepFac... See more

Daha DeepSeek'i konuşamadan Çin'den yeni bir model geldi. ByteDance Doubao-1.5-pro, DeepSeek'ten 5 kat daha uygun fiyatlı yani ChatGPT o1'in 200 katı uygun :)) 1⃣ En zorlu bilgi sorularında herkesi geçmiş. 2⃣ Kod anlamadan ve üretmede geçmiş. 3⃣ Fullstack yazılım konusunda geçmiş.___LINEBREAK... See more

Deep Research on Perplexity scores 21.1% on Humanity’s Last Exam, outperforming Gemini Thinking, o3-mini, o1, DeepSeek-R1, and other top models. We also have optimized Deep Research for speed. https://t.co/GPPFg9NbJc

"My benchmark for large language models" https://t.co/YZBuwpL0tl Nice post but even more than the 100 tests specifically, the Github code looks excellent - full-featured test evaluation framework, easy to extend with further tests and run against many LLMs. https://t.co/KnmDD1AJci E.g.... See more

🧵 Super proud to finally share this work I led last quarter - the @Databricks Domain Intelligence Benchmark Suite (DIBS)! TL;DR: Academic benchmarks ≠ real performance and domain intelligence > general capabilities for enterprise tasks. 1/3 https://t.co/n2RFxTLRaU

Destroys 4o; we have frontier models @ home I guess now https://t.co/v80G0DVHky

🔭 Analytics & Monitoring @Humanloop: Humanloop is like datadog for LLMs. They give you the tools you need to evaluate LLM apps and then take action to improve them. @Helicone_AI: The easiest way to capture data from your LLMs (Open Source). @Langfuse: Open-sourc... See more