Open models
246 models
≤3B parameters
All 37
Qwen3.5-2BQwen5.5M2.3B
MiniCPM5-2BOpenBMB986.8K2.5B
MiniCPM5-2B-DSparkOpenBMB367.1K324M
MiniCPM5-1BOpenBMB220.7K1.1B
LFM3-1.7BLiquid AI110K1.7B
LFM3-1.7B-BaseLiquid AI12.6K1.7B
3–9B parameters
All 51
NVIDIA-Nemotron-3-Nano-4B-BF16NVIDIA1.9M4.0B
granite-4.1-8bIBM Granite79.4K8.8B
granite-4.2-8bIBM Granite60.4K8.8B
Molmo2-8BAi247.6K8.7B
granite-4.2-3bIBM Granite41.1K3.7B
Spark-X2.6-4BXHToken32.9K4.1B
9–20B parameters
All 13
Ornith-1.0-9BOrnith794.5K9.4B
Qwen3-14BQwen13.1M14.8B

Qwen2.5-Coder-14B-InstructQwen1.7M14.8B

gemma-3-12b-itGoogle409.4K12.2B

phi-4Microsoft205.7K14.7B

gemma-2-9b-itGoogle148.9K9.2B
20–40B parameters
All 40
Qwen3.8-27BQwen6.9M27.8B

gemma-4-31B-itGoogle6.3M31.3B

gemma-4-26B-A4B-itGoogle5.1M25.8B

Qwen3.5-27BQwen1.7M27.8B

Qwen3.5-35B-A3BQwen1.4M36.0B
Ornith-1.0-35BOrnith1.4M35.1B
40–80B parameters
All 15
Qwen3-Coder-NextQwen347.7K79.7B

Llama-3.3-70B-InstructMeta Llama344.4K70.6B

Mixtral-8x7B-Instruct-v0.1Mistral AI325.3K46.7B

Qwen2.5-72B-InstructQwen155.6K72.7B

Llama-3.1-70B-InstructMeta Llama90.5K70.6B

Meta-Llama-3-70B-InstructMeta Llama50.5K70.6B
80–300B parameters
All 40
DeepSeek-V4-FlashDeepSeek1.2M291B

MiniMax-M2.5MiniMax1M229B
Inkling-SmallThinking Machines1M266B
Hy3Tencent987.2K299B

NVIDIA-Nemotron-3-Super-120B-A12B-BF16NVIDIA874.7K124B

Qwen3.5-122B-A10BQwen775.1K125B
>300B parameters
All 45
DeepSeek-V3.2DeepSeek9.6M685B
GLM-5.3-FlashZ.ai6.4M321B
DeepSeek-V4-Flash-0731DeepSeek2M304B
GLM-5.3Z.ai1.7M753B
DeepSeek-V4.1-FlashDeepSeek1.6M763B
InklingThinking Machines1.5M952B