-
Getting Started with Llama 3.1
The newly released Llama 3.1 series of LLMs are Meta’s “most capable models to date”. The largest 405B model is the first open source LLM to match or exceed the performance of SOTA closed-source models such as GPT-4o and Claude 3.5 Sonnet. While the 405B model is probably too big for personal computers, Meta has used it to further train and finetune smaller Llama 3 models. The results are spectacular! Compared with Llama 3 8B, the Llama 3.…
-
Mathstral: A New LLM that is Good at Math Reasoning
Today, Mistral AI released mathstral, a finetuned 7B model specifically designed for math reasoning and scientific discovery. The model has a 32k context window. The model weights are available under the Apache 2.0 license. As we have seen, leading edge LLMs, such as the GPT-4o, can solve very complex math problems. But do they have common sense? A meme that has been going around on the Internet suggests that LLMs can only pretend to solve “math Olympiad level” problems since it lacks understanding of even elementary school math.…
-
Getting Started with internlm2_5-7b-chat
The internlm2_5-7b-chat model, a new open-source model from SenseTime, introduces a 7 billion parameter base model alongside a chat model designed for practical applications. This model showcases exceptional reasoning capabilities, achieving state-of-the-art results in math reasoning tasks, outperforming competitors like Llama3 and Gemma2-9B. With a remarkable 1M context window, InternLM2.5 excels in processing extensive data, leading in long-context challenges such as LongBench. The model is also capable of tool use, integrating information from over 100 web sources, with enhanced functionalities in instruction adherence, tool selection, and reflective processes.…
-
Building a Translation Agent on LlamaEdge
By MileyFu, CNCF Ambassador, DevRel and Founding Member of WasmEdge runtime. Prof. Andrew Ng's agentic translation is a great demonstration on how to cooridnate multiple LLM “agents” to work on a single task. It allows multiple smaller LLMs (like Llama-3 or Gemma-2) to work gether and produce better results than a single large LLM (like ChatGPT). The translation agent is a great fit for LlamaEdge, which provides a lightweight, embeddable, portable, and Docker-native AI runtime for many different types of models and hardware accelerators.…
-
Getting Started with Gemma-2-9B
Google recently released Gemma 2 models in 9B and 27B Sizes, which are the latest models of its Gemma models family. According to its technical report, there will be an open sourced Gemma-2-2b model in the upcoming days. The technical report also demonstrates that the Gemma-2-9B model outperforms the Mistral-7B, Llama-3-8B, and the Gemma 1.5 models in several benchmarks. In this article, taking Gemma-2-9B as an example, we will cover…
-
Getting Started with Mistral-7B-Instruct-v0.3
The Mistral-7B-Instruct-v0.3-GGUF model is powered by the innovative GPT architecture, tailored specifically for instructional text understanding, offering unparalleled capabilities in comprehending and generating instructional content. With a vast dataset and rigorous training, Mistral-7B-Instruct-v0.3-GGUF excels in tasks ranging from parsing complex procedural instructions to generating clear and concise instructional texts across various domains. Whether it's guiding users through intricate processes or assisting educators in creating engaging educational materials, this model stands as a pinnacle in the realm of instructional NLP.…
-
Getting Started with Qwen2-7B-Instruct
Meet Qwen2-7B-Instruct, a powerhouse language model from Alibaba! It's the next generation of Qwen models, boasting serious smarts across various tasks. Compared to previous models, Qwen2-7B-Instruct blows past most open-source options and even competes with secretive proprietary models. This isn't your average language model either. Qwen2-7B-Instruct can handle massive amounts of information, crunching through text up to 131,072 tokens long. That's like tackling a whole book at once! Whether you're working with complex code, trying to solve a mind-bending math problem, or just need some serious language skills, Qwen2-7B-Instruct is ready to impress.…
-
Getting Started with Codestral-22B-v0.1
Getting Started with Codestral-22B-v0.1 The Codestral-22B-v0.1 is an advanced machine learning model designed to handle a wide array of programming tasks across over 80 programming languages, including popular ones such as Python, Java, C, C++, JavaScript, and Bash. It is specifically tailored for software development, capable of interpreting, documenting, explaining, and refactoring code. The model supports an “instruct” mode which enables it to generate code based on specific instructions, and a “Fill in the Middle” (FIM) mode that predicts missing code tokens between given code snippets.…
-
WebAssembly on Kubernetes: from containers to Wasm (part 02)
By Seven Cheng Community post by Seven Cheng | View part one here In the previous article, I gave an overview of Wasm’s features and advantages. I also explained how to run Wasm modules within container environments. In this article, I will guide you through building and deploying Wasm applications in the Cloud Native ecosystems. You’ll need: a login to Docker Hub (you can also adapt the walkthrough to use a different container image registry.…
-
Getting Started with Phi-3-mini-128k
The Phi-3-Mini-128K-Instruct is a cutting-edge model with 3.8 billion parameters, designed for lightweight yet powerful natural language processing tasks. Trained on the Phi-3 datasets, which include synthetic and filtered publicly available website data, this model prioritizes high-quality and reasoning-dense properties. It belongs to the Phi-3 family and comes in two variants: 4K and 128K, referring to the context length it can handle in tokens. Following its initial training, the model underwent a rigorous post-training process involving supervised fine-tuning and direct preference optimization.…
