Large Language Models logopeterindia.net
Knowledge Hub

Large Language Models: Frontier & Open-weight

A directory of large language models — frontier assistants and labs, open-weight model families, and leaderboards for comparing them, focused on widely used current models.

42 topics•4 categories•Updated November 2026

Frontier Assistants & Labs

14 entries

Assistants and model families from leading AI labs.

01OpenAI GPTOpenAI's GPT models behind ChatGPT and the API; the GPT-6 family spans Astra for complex reasoning and coding, Sol, Terra and the cost-efficient Luna. 02Anthropic ClaudeAnthropic's Claude assistant and model families, Fable, Opus, Sonnet and Haiku, used for agentic coding, knowledge work and enterprise applications. 03Google GeminiGoogle DeepMind's multimodal Gemini models, now the Gemini 4 generation, available in the Gemini app, Google products and the API. 04xAI GrokxAI's Grok models and assistant, aimed at coding, agentic workflows and real-time information tasks. 05Mistral AIA European AI platform offering open and commercial models, assistants and agents that enterprises can customize, fine-tune and deploy. 06DeepSeekThe AI lab behind the open-weight DeepSeek-V4 and DeepSeek-R1 models, with a chat app and an API. 07Qwen (Alibaba)Alibaba's Qwen models and chatbot, covering image and video understanding, document processing, web search and tool use. 08Kimi (Moonshot AI)Moonshot AI's Kimi assistant and models, built for agentic coding and knowledge work. 09Z.ai GLMThe Z.ai assistant, powered by the GLM model family, for building websites, writing code and handling long-horizon tasks. 10MiniMaxA multimodal model company building AI-native products used by hundreds of millions of people. 11Meta MuseMeta's Muse models, starting with Muse Spark, available through Meta AI and the Meta Model API. Muse Spark is Meta's first frontier model without open weights. 12Cohere CommandCohere's Command family of scalable language models for enterprise applications. 13Amazon NovaAmazon's family of foundation models and services, available through AWS. 14ByteDance SeedByteDance's Seed research team and its Seed model family, including the models behind the Doubao assistant.

Open-weight Model Families

14 entries

Models whose weights you can download and run.

15Meta LlamaMeta's open-weight Llama model family. Llama 4 (April 2025) is the newest Llama in Meta's Hugging Face organisation, and Meta's 2026 focus moved to its Muse models. 16Google GemmaGoogle DeepMind's family of open models, built from the research behind Gemini. 17Microsoft PhiMicrosoft's efficient small language models for generative AI applications, available in Azure AI Foundry. 18IBM GraniteIBM's family of open, trusted models for business, covering language, vision, speech and guardrail models. 19OpenAI open-weight models (gpt-oss)OpenAI's open-weight language models, which can be downloaded and run on your own infrastructure. 20NVIDIA NemotronNVIDIA's family of open models, released with open weights, training data and recipes, for building specialized reasoning agents. 21Olmo (Ai2)Ai2's fully open language model, released together with the complete model flow. 22Liquid AILiquid Foundation Models (LFM) — compact, memory-efficient models designed to run on phones, laptops and edge devices. 23SmolLM (Hugging Face)Hugging Face's SmolLM family of small, fully open language models, with training data and recipes published openly. 24Xiaomi MiMoXiaomi's open MiMo models for reasoning, coding and agentic tasks, including the MiMo-V2 Flash line. 25Tencent Hunyuan (Hy)Tencent's open Hunyuan and Hy models, from language and multimodal models to image, video and 3D generation. 26Baidu ERNIEBaidu's ERNIE model family, with open-source ERNIE models and tooling on the PaddlePaddle platform. 27Sarvam AIIndian AI company building language models for Indian languages and enterprise voice and text applications. 28TII FalconFalcon family of open models from the Technology Innovation Institute in Abu Dhabi.

Compare & Discover Models

5 entries

Leaderboards and catalogues.

Foundational Papers & Leaderboards

4 entries

Core papers, scaling laws and public evaluations of large language models.

Related Pages

5 entries

Related directories on this site.

Frequently Asked Questions

What is a large language model?

An LLM is a neural network, usually a transformer, trained on very large text corpora to predict and generate language, which lets it answer questions, write, summarise and write code.

How do I choose an LLM for an application?

Start from the task, then compare quality, price, speed, context length, licence and data-handling terms on your own examples. Leaderboards such as Arena and Artificial Analysis help shortlist models. See the LLM Evaluation Methods page on this site.

Where can I serve or run these models?

Hosted APIs from the labs and providers, cloud services and your own hardware are the main options. See the LLM Inference Platforms and Platforms for Local LLMs pages, and the LLM Hub for the full set of guides on this site.

How do I control and secure LLM use in an organisation?

Add guardrails and a gateway in the request path, test applications, and map controls to a governance framework. See the AI Agent Guardrails, AI Agent Gateways and AI Governance Frameworks pages on this site.

What does open-weight mean?

The trained model weights are published so you can run and fine-tune the model yourself, though licences vary and not all are fully open source.