Mid-size: 10 to 70 billion
The workhorse open models: Llama 2 13B, Qwen 32B, Gemma 27B and their kin. 103 models.
🇺🇸 United StatesGemma
Google DeepMind's open family of Gemini-derived language modelsGeneral · Open weights · 2024
🇺🇸 United StatesGranite 4.0
IBM's hybrid Mamba/transformer language models for enterprise workGeneral · Open weights · 2025
🇺🇸 United StatesGpt-oss-20b
OpenAI's smaller open-weight reasoning model for local and low-latency useReasoning · Open weights · 2025
🇺🇸 United StatesGemma 4
Google DeepMind's open multimodal models in five sizes, from phones to serversGeneral · Open weights
🇺🇸 United StatesPhi-3
Small Microsoft language models built to run well on phones and laptops.Small · Open weights · 2024
🇫🇷 FranceMixtral 8x7B
Sparse mixture-of-experts language model with open weightsGeneral · Open weights · 2023
🇮🇳 IndiaParam-2-17B
Multilingual MoE model for Indian languages and reasoningGeneral · Open weights
🇨🇳 ChinaDeepSeekMoE
Mixture-of-experts language model architecture by DeepSeek AIGeneral · Open weights · 2024
🇰🇷 South KoreaKanana-2
Open-source multilingual model for agentic AI and reasoning.General · Open weights
🇨🇳 ChinaKimi Linear
Hybrid linear attention model with 48 billion parametersGeneral · Open weights · 2025
🇮🇳 IndiaSarvam-30B
Open-weight MoE model for Indian languages and codingGeneral · Open weights
🇺🇸 United StatesNemotron 3 Nano
NVIDIA's 30B hybrid Mamba-Transformer MoE model for reasoning and agentsGeneral · Conditional licence · 2025