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

Mixture of experts

Mixtral, DeepSeek-V3, Kimi K2: many expert networks, a few used for each token. 112 models on EveryLLM so far.

A mixture of experts (MoE) model replaces each dense feed-forward block with many smaller expert networks and a router that sends each token to only a few of them. The model can hold a very large number of parameters while using only a fraction for any one token, so it learns more without costing proportionally more to run. Mixtral, DeepSeek-V3, Qwen3, Llama 4, gpt-oss and Kimi K2 are all built this way.

Total and active parameters

An MoE has two sizes. DeepSeek-V3 (December 2024) holds 671 billion parameters but activates 37 billion for each token. Mixtral 8x7B (December 2023) has eight experts per layer, uses two for each token, and has 46.7 billion parameters in total with 12.9 billion active. Kimi K2 (July 2025) reaches one trillion in total with 32 billion active. All the experts must still sit in memory, so MoE models need large servers even when they are fast.

A long history

The idea dates to 1991, and Google researchers led by Noam Shazeer revived it for language in 2017 with a sparsely gated layer of thousands of experts. Google's Switch Transformer reached 1.6 trillion parameters in 2021. Open MoE models took off with Mixtral, then Databricks' DBRX (March 2024, 132 billion total, 36 billion active), and became the norm for large Chinese open models. Training them is harder, because routers must spread tokens evenly across experts.

Famous mixture of experts models

Further reading

fnK-EXAONE 🇰🇷 South KoreaK-EXAONE LG AI Research multilingual Mixture-of-Experts modelGeneral · Open weights Granite 3.0 🇺🇸 United StatesGranite 3.0 IBM's third-generation open enterprise language models, from 2B to 8B parametersGeneral · Open weights · 2024 imgDeepSeek-OCR WorldDeepSeek-OCR Optical character recognition model using visual compressionMultimodal · Open weights · 2025 Grok-1 🇺🇸 United StatesGrok-1 xAI's 314 billion parameter open-weights mixture-of-experts base modelGeneral · Open weights · 2024 Sarvam-105B 🇮🇳 IndiaSarvam-105B Open-weight MoE model for Indian languages and reasoningGeneral · Open weights imgvidfnKimi K2.5 🇨🇳 ChinaKimi K2.5 Open-source multimodal agentic model with vision and reasoning.Multimodal · Open weights Hunyuan-Large 🇨🇳 ChinaHunyuan-Large Open-source mixture of experts model with 389 billion parametersGeneral · Open weights · 2024 Param-1 🇮🇳 IndiaParam-1 Bilingual and multilingual language models from India.General · Conditional licence fnMiniMax-M2 🇨🇳 ChinaMiniMax-M2 Open-source MoE model for coding and agentsCode · Open weights fnNemotron 3 Nano 🇺🇸 United StatesNemotron 3 Nano NVIDIA's 30B hybrid Mamba-Transformer MoE model for reasoning and agentsGeneral · Conditional licence · 2025 imgvidfnKimi K3 🇨🇳 ChinaKimi K3 Open-weight multimodal agentic model with 2.8T parametersGeneral · Open weights · 2026 DeepSeek-V2 🇨🇳 ChinaDeepSeek-V2 Efficient mixture-of-experts language model with 236 billion parametersGeneral · Open weights · 2024 OLMoE 🇺🇸 United StatesOLMoE Open mixture-of-experts language model with 1 billion active parametersGeneral · Fully open · 2024 imgvidfnStep-5-Preview WorldStep-5-Preview Agentic model with 600B parameters and 1M token contextGeneral · Open weights · 2026 DeepSeek-R1 🇨🇳 ChinaDeepSeek-R1 Open-weight reasoning model from ChinaReasoning · Open weights · 2024 Jamba 🇮🇱 IsraelJamba Hybrid open-weights language model with long contextGeneral · Open weights · 2024

The weekly digest

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