Can I run SmolLM3-3B on an RTX 5060?

Yes, just

Its published weights (6.15 GB) fit in the 7.20 GB an RTX 5060 can use, with room for ≈ 14K tokens of context.

Full analysis of SmolLM3-3B Everything that runs on RTX 5060 Get the weights on Hugging Face

hf download HuggingFaceTB/SmolLM3-3B

Files and context on RTX 5060

Each file of SmolLM3-3B (and of GGUF versions the site has analyzed), and how much context fits next to it, with the context cache in 16-bit or 8-bit.

File	Size	On RTX 5060	Context (16-bit)	Context (8-bit)
As published	6.15 GB	Tight	14K	28K
4-bit (estimate)	1.73 GB	Fits	64K	64K

SmolLM3-3B on other hardware

NVIDIA H100Yes
RTX 5090Yes
RTX 4090Yes
RTX 3090Yes
RTX 5060 Ti 16 GBYes
RTX 3060 12 GBYes
RTX 5060Yes, just
RTX 4060Yes, just
Mac with 32 GBYes
Mac with 64 GBYes
Mac with 128 GBYes

Popular models that run on RTX 5060

Qwen3-0.6BQwen · 752M
gpt2openai-community · 137M
Qwen2.5-0.5B-InstructQwen · 494M
Qwen2.5-1.5B-InstructQwen · 1.54B
Ornith-1.5-9B-GGUFornith-ai · 9.20B
Qwen2.5-3B-InstructQwen · 3.09B
Qwen3.5-2BQwen · 2.27B
Qwen3-1.7BQwen · 2.03B

Assumes 90% of the card's memory is usable and at least 4K tokens of context for one conversation, with the context cache in 16-bit unless noted. Fits leaves at least 20% of that memory for context; Tight leaves less.