News
ReleasesOVHcloud unveils Olala mini, its first in-house LLMChinaA 9B model beats a 1T model on FinQA, trained with Crusoe Managed RLReleasesThe internal metaphysics of an LLMReleasesMapping Language Model Internal Representations to Physical Laws and Scientific ReasoningResearchCan China keep rogue AI in check? Meet the start-up testing the top modelsFrontierMicrosoft aims to challenge OpenAI, Anthropic with cheaper, specialized AI modelsOpen modelsMicrosoft rolls out new first-party models and updates to agentic AI continuous improvementFrontierMicrosoft unveils AI models for coding, cybersecurity as it takes on Anthropic, OpenAIFrontierGoogle is making it easier to opt out of Gemini AI model training – here’s howFrontierOpenAI Unveils New Astra AI Model After Scrapping Prior ReleaseReleasesOVHcloud unveils Olala mini, its first in-house LLMChinaA 9B model beats a 1T model on FinQA, trained with Crusoe Managed RLReleasesThe internal metaphysics of an LLMReleasesMapping Language Model Internal Representations to Physical Laws and Scientific ReasoningResearchCan China keep rogue AI in check? Meet the start-up testing the top modelsFrontierMicrosoft aims to challenge OpenAI, Anthropic with cheaper, specialized AI modelsOpen modelsMicrosoft rolls out new first-party models and updates to agentic AI continuous improvementFrontierMicrosoft unveils AI models for coding, cybersecurity as it takes on Anthropic, OpenAIFrontierGoogle is making it easier to opt out of Gemini AI model training – here’s howFrontierOpenAI Unveils New Astra AI Model After Scrapping Prior Release
EveryLLM

Dense decoder transformers

The GPT design: one stack of transformer layers predicting the next token. 308 models on EveryLLM so far.

The dense decoder transformer is the design behind GPT and most chatbots. It is a single stack of identical transformer layers that reads the tokens so far and predicts the next one, with a mask that stops each position from looking ahead. Dense means every parameter is used for every token, unlike in a mixture of experts. GPT-3, PaLM, Llama 3, Chinchilla, Gemma, Phi and OLMo are dense decoders.

Inside each layer

Each layer has two parts. Self-attention lets every token weigh every earlier token and gather what it needs; a feed-forward network then transforms each token's vector on its own. The transformer was introduced by Google researchers in the June 2017 paper Attention Is All You Need, for translation, and OpenAI's GPT-1 (2018) kept only its decoder half. Llama 3.1 405B, released in July 2024, stacks 126 such layers.

Why it won

A decoder trains on any text without labels, because the next token is always the answer to predict, and it scales smoothly: scaling laws published by OpenAI in 2020 and DeepMind in 2022 showed the error falling predictably with more parameters, data and compute. Its weak point is cost. Attention grows with the square of the context length, and every parameter must be read for every token, which is why mixtures of experts became popular.

Famous dense decoder models

Still to be written: OLMo (language model).

Further reading

imgwavKimi 🇨🇳 ChinaKimi A Chinese chatbot and large language model family by Moonshot AI.Assistant · API only · 2023 fnDeepSeek 🇨🇳 ChinaDeepSeek A Chinese series of large language models and a free chatbot appGeneral · API only · 2025 imgwavvidGemini 🇺🇸 United StatesGemini Google DeepMind's multimodal LLM family, from Nano to UltraGeneral · API only · 2023 S1 🇺🇸 United StatesS1 A reasoning model using test-time scaling and budget forcingReasoning · Fully open · 2025 imgMolmo 🇺🇸 United StatesMolmo Open vision-language model family by Allen Institute for AIMultimodal · Fully open · 2024 Taiwan-LLM 🇹🇼 TaiwanTaiwan-LLM Open language model for Traditional Chinese in TaiwanRegional · Fully open · 2023 Poro 🇫🇮 FinlandPoro Open 34B parameter model for Finnish, English and codeGeneral · Fully open · 2024 Sarvam-Translate 🇮🇳 IndiaSarvam-Translate Document-level translation model for 22 Indian languages.Domain · Fully open Viking 🇫🇮 FinlandViking Open source language model for Nordic languages and codeGeneral · Fully open · 2023 imgLLaVA 🇺🇸 United StatesLLaVA Open-source multimodal chatbot connecting vision encoders to language models.Multimodal · Fully open · 2023 Megatron-LM 🇺🇸 United StatesMegatron-LM Model-parallel training techniques for transformer language models with billions of parametersResearch · Fully open · 2019 AceGPT 🇨🇳 ChinaAceGPT An Arabic-focused large language model designed for cultural alignment and local values.Regional · Fully open · 2023

The weekly digest

Every Monday: the model of the day, new pages on EveryLLM, and the conferences and anniversaries of the week ahead.