Cursor AI—an online AI Code Editor, recently published (and removed) a blog post on a new model and inference method for high-accuracy full-file edits at 1000 tokens/second. The blog post is no longer accessible, but we are reposting the methodology they had published here for the readers. This is the link to the original Cursor […]
Author: Sushant Babbar
Meta has recently introduced the Llama 3.2 version to the Llama LLM family, which follows the release of Llama 3.1 405B – a model lauded for being one of the most budget-friendly and advanced open-source foundation models. This latest offering by Meta comes in 1B and 3B sizes that are multilingual text-only and 11B and […]
Anthropic has recently introduced ‘Contextual Retrieval’ for Claude, a method that they believe dramatically improves the retrieval step in Retrieval-Augmented Generation (RAG). Following the launch of Claude for Enterprise and prompt caching, which helps the LLM models cache, people are already excited about its potential for coding tasks. This new feature enhances how AI helps […]
We’re halfway through 2025, and AI code generation remains one of the top AI applications. As such, we have many platforms providing advanced AI code generation capabilities and features. The most popular among these is ChatGPT, but is ChatGPT the best option for AI code generation and development? Indeed, many better alternatives are available, and […]
2024 has proven to be a solid year for AI code generation. The DeepSeek models, often overlooked in comparison to GPT-4o and Claude 3.5 Sonnet, have gained decent momentum in the past few months. And with the recent announcement of DeepSeek 2.5, an upgraded version that combines DeepSeek-V2-Chat and DeepSeek-Coder-V2-Instruct, the momentum has peaked. DeepSeek-V2.5 […]
Google has recently announced DataGemma, a pair of instruction-tuned models engineered for better accuracy. It’s interesting for two main reasons: 1. It’s trained on vast real-world data to mitigate the challenge of hallucinations. 2. It’s open-source. And with the recent announcement of OpenAI o1—also designed with accuracy and reasoning in mind—people have started to draw […]