LLM Explorer Blog

The LLM Explorer Rank (Score) is a comprehensive metric for dynamic evaluation of language models. It combines factors like popularity, recency, and expert ratings to provide a balanced assessment. The system uses normalized weights, logarithmic scaling, and a recency boost to ensure fair...

To begin with, AI professionals are really excited to see Mistral's ability to develop and release high-quality language models rapidly. The AI community is particularly glad to see more competition in the space of small language models that can run efficiently on consumer-grade GPUs, like the...

The release of Mathstral, a new language model specializing in STEM subjects and mathematical reasoning, started a discussion about the future of math-focused AI. AI professionals, enthusiasts, and researchers talked about whether it's better to have separate tools for math or to build math skills...

Mamba-Codestral-7B-v0.1, a new language model for code generation from the Mistral AI team, has become one of the most notable recent releases in the AI community.  The Mamba architecture offers several key advantages. It can process long texts efficiently, with generation time remaining...

This week's one of the top-trending models on LLM Explorer is Tiger-Gemma-9B-v1, created by TheDrummer: The model's name is a tribute to a street cat 🐈, with pictures available in the model card. Tiger-Gemma-9B-v1 is a modified version of Gemma 9B with fewer restrictions. It's important to note...

Last week, the Numina team released NuminaMath 7B TIR, a new math-focused language model. It quickly reached the top position in our "Top-Trending LLMs over the Last Week" ranking. For context, our ranking is based on the number of downloads and likes, using data from Hugging Face and LLM...

Last week, Hugging Face released the Open LLM Leaderboard v2. This update addresses the shortcomings of the first version. The new leaderboard includes stricter benchmarks, improved evaluation methods, and a more balanced scoring system. These enhancements aim to give the AI community a clearer...

Last week, Google released Gemma 2, the latest addition to its family of state-of-the-art open LLMs. It comes in two sizes: 9 billion and 27 billion parameters, with both base (pre-trained) and instruction-tuned versions. Gemma 2 has the same permissive license as the first iteration, allowing...

  Looking at our list of Top-trending models (ranked by downloads and likes based on information from Hugging Face and LLM Explorer), we can see that LLMs from the Qwen2 family have remained popular among the AI community over the past week.   A nice experience was shared within the AI...

A week ago, the Intento team (a machine translation and multilingual generative AI platform for global enterprise companies) published its 8th annual State of Machine Translation report. The work analyzes 52 MT engines and LLMs across 11 language pairs and 9 content domains. The MT systems and...

When it comes to selecting the right language model for your frequent tasks, there's nothing like diving into a virtual discussion with AI enthusiasts. Here's a glimpse into which models are favored by professionals for different needs. This is just a friendly chat, not an official guide, but it's...

We enjoy hearing AI enthusiasts compare various models, and today, we're exploring discussions about C4AI Command R+ from CohereForAI 😊. Model Summary: C4AI Command R+ is a highly advanced 104 billion parameter model by Cohere and Cohere For AI. It excels in automating complex tasks using...

Mistral AI has recently released Codestral 22B V0.1, an open-weight generative AI model designed specifically for code generation tasks across more than 80 programming languages, including Python, Java, and JavaScript. Codestral is touted to not only complete functions and write tests but also...

Last week's news about California’s new AI bill sparked many discussions in the AI community about the future of AI development. The bill aims to ensure responsible AI development but imposes significant restrictions on large language models. Critics argue that it introduces numerous...

Mistral AI has released the Mistral-7B-Instruct-v0.3 model (license: apache-2.0), and it comes with a significant feature: function calling. This is remarkable because it's available in a medium-sized model with 7 billion parameters, making advanced capabilities more accessible. Why Function...