Best Large Language Models in the UK - Page 16

Gemini 4 Flash, reportedly codenamed Hydrogen, is an anticipated multimodal artificial intelligence model in Google DeepMind's Gemini family. The model is expected to emphasize fast response times, efficient inference, and advanced reasoning for everyday AI applications. Potential capabilities include natural language understanding, code generation, document analysis, visual reasoning, and automated task execution. Based on Google's existing Flash models, Hydrogen may support tool calling, structured outputs, long-context processing, and integration with external applications. Potential applications include AI assistants, customer service automation, software development, content generation, and enterprise productivity workflows. Google has not officially confirmed the Hydrogen codename, model specifications, pricing, or release date.

Llama (Large Language Model Meta AI) is a state-of-the-art foundational large language model designed to help researchers advance their work in this subfield of AI. Smaller, more performant models such as Llama enable others in the research community who don’t have access to large amounts of infrastructure to study these models, further democratizing access in this important, fast-changing field. Training smaller foundation models like Llama is desirable in the large language model space because it requires far less computing power and resources to test new approaches, validate others’ work, and explore new use cases. Foundation models train on a large set of unlabeled data, which makes them ideal for fine-tuning for a variety of tasks. We are making Llama available at several sizes (7B, 13B, 33B, and 65B parameters) and also sharing a Llama model card that details how we built the model in keeping with our approach to Responsible AI practices.

Large language models, which are often trained for hundreds of thousands of compute days, have shown remarkable capabilities for zero- and few-shot learning. Given their computational cost, these models are difficult to replicate without significant capital. For the few that are available through APIs, no access is granted to the full model weights, making them difficult to study. We present Open Pre-trained Transformers (OPT), a suite of decoder-only pre-trained transformers ranging from 125M to 175B parameters, which we aim to fully and responsibly share with interested researchers. We show that OPT-175B is comparable to GPT-3, while requiring only 1/7th the carbon footprint to develop. We are also releasing our logbook detailing the infrastructure challenges we faced, along with code for experimenting with all of the released models.

With T5, we propose reframing all NLP tasks into a unified text-to-text-format where the input and output are always text strings, in contrast to BERT-style models that can only output either a class label or a span of the input. Our text-to-text framework allows us to use the same model, loss function, and hyperparameters on any NLP task, including machine translation, document summarization, question answering, and classification tasks (e.g., sentiment analysis). We can even apply T5 to regression tasks by training it to predict the string representation of a number instead of the number itself.

PanGu-α is developed under the MindSpore and trained on a cluster of 2048 Ascend 910 AI processors. The training parallelism strategy is implemented based on MindSpore Auto-parallel, which composes five parallelism dimensions to scale the training task to 2048 processors efficiently, including data parallelism, op-level model parallelism, pipeline model parallelism, optimizer model parallelism and rematerialization. To enhance the generalization ability of PanGu-α, we collect 1.1TB high-quality Chinese data from a wide range of domains to pretrain the model. We empirically test the generation ability of PanGu-α in various scenarios including text summarization, question answering, dialogue generation, etc. Moreover, we investigate the effect of model scales on the few-shot performances across a broad range of Chinese NLP tasks. The experimental results demonstrate the superior capabilities of PanGu-α in performing various tasks under few-shot or zero-shot settings.

Megatron-Turing Natural Language Generation model (MT-NLG), is the largest and the most powerful monolithic transformer English language model with 530 billion parameters. This 105-layer, transformer-based MT-NLG improves upon the prior state-of-the-art models in zero-, one-, and few-shot settings. It demonstrates unmatched accuracy in a broad set of natural language tasks such as, Completion prediction, Reading comprehension, Commonsense reasoning, Natural language inferences, Word sense disambiguation, etc. With the intent of accelerating research on the largest English language model till date and enabling customers to experiment, employ and apply such a large language model on downstream language tasks - NVIDIA is pleased to announce an Early Access program for its managed API service to MT-NLG mode.

Information overload is a major obstacle to scientific progress. The explosive growth in scientific literature and data has made it ever harder to discover useful insights in a large mass of information. Today scientific knowledge is accessed through search engines, but they are unable to organize scientific knowledge alone. Galactica is a large language model that can store, combine and reason about scientific knowledge. We train on a large scientific corpus of papers, reference material, knowledge bases and many other sources. We outperform existing models on a range of scientific tasks. On technical knowledge probes such as LaTeX equations, Galactica outperforms the latest GPT-3 by 68.2% versus 49.0%. Galactica also performs well on reasoning, outperforming Chinchilla on mathematical MMLU by 41.3% to 35.7%, and PaLM 540B on MATH with a score of 20.4% versus 8.8%.

Significant advancements in the field of natural language processing, understanding, and generation have been achieved through the expansion of large language models. This study introduces a system which utilizes Ascend 910 AI processors and the MindSpore framework to train a language model with over a trillion parameters, specifically 1.085T, named PanGu-{\Sigma}. This model, which builds upon the foundation laid by PanGu-{\alpha}, takes the traditionally dense Transformer model and transforms it into a sparse one using a concept known as Random Routed Experts (RRE). The model was efficiently trained on a dataset of 329 billion tokens using a technique called Expert Computation and Storage Separation (ECSS), leading to a 6.3-fold increase in training throughput via heterogeneous computing. Experimentation indicates that PanGu-{\Sigma} sets a new standard in zero-shot learning for various downstream Chinese NLP tasks.

OpenELM is an open-source language model family developed by Apple. It uses a layer-wise scaling strategy to efficiently allocate parameters within each layer of the transformer model, leading to enhanced accuracy compared to existing open language models of similar size. OpenELM is trained on publicly available datasets and achieves state-of-the-art performance for its size.

LTM-2-mini is a 100M token context model: LTM-2-mini. 100M tokens equals ~10 million lines of code or ~750 novels. For each decoded token, LTM-2-mini’s sequence-dimension algorithm is roughly 1000x cheaper than the attention mechanism in Llama 3.1 405B1 for a 100M token context window. The contrast in memory requirements is even larger – running Llama 3.1 405B with a 100M token context requires 638 H100s per user just to store a single 100M token KV cache.2 In contrast, LTM requires a small fraction of a single H100’s HBM per user for the same context.

The o3-mini-high model from OpenAI advances AI reasoning by refining deep problem-solving in coding, mathematics, and complex tasks. It features adaptive thinking time with adjustable reasoning modes (low, medium, high) to optimize performance based on task complexity. Outperforming the o1 series by 200 Elo points on Codeforces, it delivers high efficiency at a lower cost while maintaining speed and accuracy. As part of the o3 family, it pushes AI problem-solving boundaries while remaining accessible, offering a free tier and expanded limits for Plus subscribers.

Contextual AI introduces its Grounded Language Model (GLM), engineered specifically to minimize hallucinations and deliver highly accurate, source-based responses for retrieval-augmented generation (RAG) and agentic applications. The GLM prioritizes faithfulness to the provided data, ensuring responses are grounded in specific knowledge sources and backed by inline citations. With state-of-the-art performance on the FACTS groundedness benchmark, the GLM outperforms other foundation models in scenarios requiring high accuracy and reliability. The model is designed for enterprise use cases like customer service, finance, and engineering, where trustworthy and precise responses are critical to minimizing risks and improving decision-making.

ERNIE 4.5 Turbo, unveiled by Baidu at the 2025 Baidu Create conference, is a cutting-edge AI model designed to handle a variety of data inputs, including text, images, audio, and video. It offers powerful multimodal processing capabilities that enable it to perform complex tasks across industries such as customer support automation, content creation, and data analysis. With enhanced reasoning abilities and reduced hallucinations, ERNIE 4.5 Turbo ensures that businesses can achieve higher accuracy and reliability in AI-driven processes. Additionally, this model is priced at just 1% of GPT-4.5’s cost, making it a highly cost-effective alternative for enterprises looking for top-tier AI performance.

ERNIE X1.1 is Baidu’s upgraded reasoning model that delivers major improvements over its predecessor. It achieves 34.8% higher factual accuracy, 12.5% better instruction following, and 9.6% stronger agentic capabilities compared to ERNIE X1. In benchmark testing, it surpasses DeepSeek R1-0528 and performs on par with GPT-5 and Gemini 2.5 Pro. Built on the foundation of ERNIE 4.5, it has been enhanced with extensive mid-training and post-training, including reinforcement learning. The model is available through ERNIE Bot, the Wenxiaoyan app, and Baidu’s Qianfan MaaS platform via API. These upgrades are designed to reduce hallucinations, improve reliability, and strengthen real-world AI task performance.

ERNIE 5.0 is a next-generation conversational AI platform developed by Baidu, designed to deliver natural, human-like interactions across multiple domains. Built on Baidu’s Enhanced Representation through Knowledge Integration (ERNIE) framework, it fuses advanced natural language processing (NLP) with deep contextual understanding. The model supports multimodal capabilities, allowing it to process and generate text, images, and voice seamlessly. ERNIE 5.0’s refined contextual awareness enables it to handle complex conversations with greater precision and nuance. Its applications span customer service, content generation, and enterprise automation, enhancing both user engagement and productivity. With its robust architecture, ERNIE 5.0 represents a major step forward in Baidu’s pursuit of intelligent, knowledge-driven AI systems.

Grok 4.20 is an advanced artificial intelligence model developed by xAI to elevate reasoning and natural language understanding. Built on the high-performance Colossus supercomputer, it is engineered for speed, scale, and accuracy. Grok 4.20 processes multimodal inputs such as text and images, with video support planned for future releases. The model excels in scientific, technical, and linguistic tasks, delivering highly precise and context-aware responses. Its architecture supports deep reasoning and sophisticated problem-solving capabilities. Enhanced moderation improves output reliability and reduces bias compared to earlier versions. Overall, Grok 4.20 represents a significant step toward more human-like AI reasoning and interpretation.

Grok 5 is an anticipated next-generation artificial intelligence model from xAI, designed to advance the capabilities of the Grok family of large language models. The model is expected to focus on complex reasoning, software engineering, multimodal understanding, and autonomous AI agent workflows. Potential capabilities include code generation, mathematical problem-solving, scientific analysis, document processing, and advanced information retrieval. Building on previous Grok models, it may support tool calling, extended context processing, real-time information access, and multistep task execution. Grok 5 could be used for AI assistants, software development, enterprise automation, technical research, and productivity applications. The model has not been officially released, and its architecture, context window, performance benchmarks, pricing, and availability remain unconfirmed.

Chinchilla is a large language model. Chinchilla uses the same compute budget as Gopher but with 70B parameters and 4× more more data. Chinchilla uniformly and significantly outperforms Gopher (280B), GPT-3 (175B), Jurassic-1 (178B), and Megatron-Turing NLG (530B) on a large range of downstream evaluation tasks. This also means that Chinchilla uses substantially less compute for fine-tuning and inference, greatly facilitating downstream usage. As a highlight, Chinchilla reaches a state-of-the-art average accuracy of 67.5% on the MMLU benchmark, greater than a 7% improvement over Gopher.

OpenAI Bel is the reported codename of an unreleased artificial intelligence foundation model associated with OpenAI. Unofficial reports describe it as a large-scale pretraining project intended to support future generations of AI models. The project has been linked in speculative reporting to OpenAI's Astra model family and potential successors to GPT-6. Some sources claim that Bel contains more than 10 trillion parameters, although this figure has not been independently verified. OpenAI has not published official technical specifications, benchmark results, or documentation confirming the project's capabilities. Bel is not currently available as a publicly documented model through ChatGPT or the OpenAI API.