Blog
-
LLama 3 – 400B
The AI research community has been abuzz with the recent release of Llama 3, a series of language models that promise to revolutionize the field of natural language processing. The initial release of Llama 3 8B and 70B models marked the beginning of an exciting journey, with even more impressive models on the horizon. In…
-
LlamaIndex on Vertex AI
LlamaIndex Team excited to partner with the Vertex AI team (@googlecloud) to feature a brand-new RAG API on Vertex, powered by @llama_index advanced modules that enable e2e indexing, embedding, retrieval, and generation. It is simultaneously easy to setup and use, while providing developers programmatic flexibility to connect a range of data sources (local, GCS, GDrive)…
-
Building JavaScript agents in LlamaIndex.TS
The ultimate guide to building agents in TypeScript is here! This guide takes you step-by-step through: What is an Agent? In LlamaIndex, an agent is a semi-autonomous piece of software powered by an LLM that is given a task and executes a series of steps towards solving that task. It is given a set of…
-
Optimizing RAG with LLamaIndex
A cool trick you can use to improve retrieval performance in your RAG pipelines is fine-tune the embedding model (bi-encoder) based on labels from a cross-encoder 💡 Cross-encoders are crucial for reranking but are way too slow for retrieving over large numbers of documents. This fine-tuning technique gives you all the speed advantages of direct…
-
Run LLama 3 on iPhone 15 Pro
In addition to other improvements, current release enables running Meta Llama 2 7B efficiently on devices like the iPhone 15 Pro, Samsung Galaxy S24 and other edge devices — it also includes early support for Llama 3 8B. More details on ExecuTorch Alpha below. ExecuTorch ExecuTorch Alpha is focused on deploying large language models and…
-
LLama 3 vs GPT-4
Llama 3 is a cutting-edge large language model introduced by Meta AI on April 18, 2024. This model family offers three sizes: 8B, 70B, and 400B. The 8B and 70B versions are available to the public, whereas the 400B version is currently undergoing training. Llama 3 boasts benchmark scores that match or surpass those of…
-
Llama-3 Is Not Really Censored
It turns out that Llama-3, right out of the box, is not heavily censored. In the release blog post, Meta indicated that we should expect fewer prompt refusals, and this appears to be accurate. For example, if you were to ask the Llama-3 70 billion model to tell you a joke about women or men,…
-
LLama 3 on Groq
Okay, so this is the actual speed of generation, and we’re achieving more than 800 tokens per second, which is unprecedented. Since the release of LLama 3 earlier this morning, numerous companies have begun integrating this technology into their platforms. One particularly exciting development is its integration with Groq Cloud, which boasts the fastest inference…
-
LLama 3 is HERE
Today marks the exhilarating launch of LLama 3! In this blog post, we’ll delve into the announcement of LLama 3, exploring what’s new and different about this latest model. If you’re passionate about AI, make sure to subscribe to receive more fantastic content. Launch Details and Initial Impressions Just a few minutes ago, we witnessed…