SmolLM3 models
3 SmolLM3 models from Hugging Face on Hugging Face, from 3.1B to 3.1B parameters, published by Hugging Face in BF16, with a 64K-token context. The smallest official model, SmolLM3-3B, needs about 6.9 GB of VRAM at its published precision; the cheapest live fit is RTX 3070 at $0.088/hr.
Part of the SmolLM series · Previous generation: SmolLM2
Pick a size
One row per official SmolLM3 size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| SmolLM3-3B-Base | 3.1B | 6.9 GB | 3.4 GB | 1.7 GB | RTX 3070 | $0.088/hr | 2.3 GB |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (2)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| SmolLM3-3B | 3.1B | BF16 | 6.9 GB | RTX 3070 | $0.088/hr |
| SmolLM3-3B-Base | 3.1B | BF16 | 6.9 GB | RTX 3070 | $0.088/hr |
Fine-tunes and community models (1)
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
Run and fine-tune SmolLM3
- How to run SmolLM3: engine support, chat template and context flags, with a serving command sized per model.
- How to fine-tune SmolLM3: LoRA and QLoRA memory per model, and the license limits on derivatives.