Google Unveils Gemini 2.5 Ultra With 1M Tokens
Google DeepMind launches Gemini 2.5 Ultra, its most powerful AI model yet, featuring a million-token context window and …
🏷️ Large Language Models
104 articles about 'Large Language Models'
Rakuten Deploys Custom LLMs to Power E-Commerce Search
Japanese tech giant Rakuten integrates proprietary large language models into its e-commerce search engine, aiming to re…
Naver HyperCLOVA X Pushes Into Southeast Asia
South Korea's Naver expands HyperCLOVA X with multilingual AI capabilities targeting Southeast Asian markets.
Cohere Launches Command R+ for Enterprise RAG
Cohere releases Command R+, a powerful LLM optimized for enterprise RAG workflows, grounded citations, and multilingual …
Claude 4 Shatters Graduate-Level AI Benchmarks
Anthropic's Claude 4 sets new records on GPQA and other graduate-level evaluations, outperforming GPT-4o and Gemini Ultr…
OpenAI Launches GPT-5 Turbo With Native Reasoning
OpenAI unveils GPT-5 Turbo, featuring built-in chain-of-thought reasoning, 1M token context, and up to 3x benchmark gain…
LG AI Research Launches EXAONE 4.0 Models
LG AI Research unveils EXAONE 4.0, its latest enterprise-focused foundation model series targeting industrial and busine…
Cohere Launches Command A for RAG Workloads
Cohere releases Command A, a new language model purpose-built for retrieval-augmented generation with enterprise-grade a…
OpenAI Launches GPT-5.5 Instant as ChatGPT Default
OpenAI replaces GPT-3.5 Instant with GPT-5.5 Instant as ChatGPT's new default model, bringing major performance upgrades…
Anthropic Launches Claude 4.5 Sonnet With Math Proofs
Anthropic releases Claude 4.5 Sonnet featuring breakthrough mathematical proof generation that outperforms GPT-4o and Ge…
GPT-5 Turbo Sets Records in Multi-Step Reasoning
OpenAI's GPT-5 Turbo achieves breakthrough scores on complex reasoning benchmarks, outpacing rivals by significant margi…
Meta Launches Llama 4 Maverick With 400B Parameters
Meta releases Llama 4 Maverick, a 400B-parameter open-weight model using mixture-of-experts architecture, challenging pr…