News
EveryLLM

English first

Trained mostly on English. 206 models on EveryLLM so far.

English-first models were trained mostly on English text, because English dominates the web, books, code comments and the research papers that set the agenda. Most early language models, from BERT and GPT-2 to Chinchilla and the first Llama, fall in this group, and many small and research models still do. They often handle other languages, but with clearly lower quality.

Where the data came from

BERT (2018) learned from English Wikipedia and a large set of unpublished books. GPT-2 (2019) used WebText, about 40 GB of text from 8 million web pages linked from Reddit. EleutherAI's Pile (2020), an 825 GiB collection weighted towards English academic papers, books and code, trained GPT-J and Pythia. Common Crawl, a free archive of the web gathered since 2008, supplies the bulk of most training sets.

Why it still matters

English-first training shapes what models know and how they speak: they may default to American spelling, law and culture, and do worse on questions about other places. Many benchmarks exist only in English. Labs now add far more non-English data, but English remains the largest share for most general models, and research models built to study training, such as OLMo, often stay English to keep the experiment clean.

Famous english first models

Still to be written: OLMo (language model).

Further reading

imgfnGrok 馃嚭馃嚫 United StatesGrok A chatbot and model line from SpaceX Super Intelligence, launched in 2023.General 路 API only 路 2023 imgLlama 馃嚭馃嚫 United StatesLlama Meta AI's family of open-weight language models, from 1B to 2T parametersGeneral 路 Conditional licence 路 2023 BLOOM 馃嚝馃嚪 FranceBLOOM A 176-billion-parameter multilingual text model built openly by a volunteer research workshop.General 路 Conditional licence 路 2022 imgfnPhi 馃嚭馃嚫 United StatesPhi Microsoft's family of small, open-weight language models that run on everyday devices.Small 路 Open weights 路 2023 DeepSeek LLM 67B 馃嚚馃嚦 ChinaDeepSeek LLM 67B A 67-billion-parameter open-source model trained on English and Chinese text.General 路 Conditional licence 路 2024 Poro 2 馃嚝馃嚠 FinlandPoro 2 Finnish and English instruction-following model based on Llama 3.1General 路 Conditional licence imgGrok-2 馃嚭馃嚫 United StatesGrok-2 xAI's 2024 chat and reasoning model, with a smaller sibling, Grok-2 miniGeneral 路 Conditional licence 路 2024 Krutrim-1 馃嚠馃嚦 IndiaKrutrim-1 Multilingual language model for Indian languagesGeneral 路 Conditional licence 路 2024 OPT 馃嚭馃嚫 United StatesOPT Meta AI's family of open decoder-only language models from 125M to 175B parametersGeneral 路 Conditional licence 路 2022 Llama 3 馃嚭馃嚫 United StatesLlama 3 Meta's open-weight language model family in 8B and 70B sizesGeneral 路 Conditional licence 路 2024 Qwen2 馃嚚馃嚦 ChinaQwen2 Open-weight language model series from AlibabaGeneral 路 Open weights 路 2024 Param-1 馃嚠馃嚦 IndiaParam-1 Bilingual and multilingual language models from India.General 路 Conditional licence Llama-3-ELYZA-JP-8B 馃嚡馃嚨 JapanLlama-3-ELYZA-JP-8B Japanese-tuned Llama 3 model by ELYZAGeneral 路 Conditional licence fnEXAONE 4.0 馃嚢馃嚪 South KoreaEXAONE 4.0 Unified language model with reasoning and non-reasoning modesGeneral 路 Conditional licence 路 2025 Nemotron-4 340B 馃嚭馃嚫 United StatesNemotron-4 340B NVIDIA's 340B open-access model family for chat and synthetic data generationGeneral 路 Conditional licence 路 2024 fnLlama-3.1-Nemotron-Ultra-253B 馃嚭馃嚫 United StatesLlama-3.1-Nemotron-Ultra-253B A 253B reasoning model derived from Llama-3.1-405B-InstructReasoning 路 Conditional licence

The weekly digest

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