SmolVLM models
4 SmolVLM models from Hugging Face on Hugging Face, from 256M to 2.2B parameters, published by Hugging Face in BF16 and F32, with a 8K-token context. The smallest official model, SmolVLM-256M-Instruct, needs about 0.6 GB of VRAM at its published precision; the cheapest live fit is RTX 3070 at $0.088/hr.
Pick a size
One row per official SmolVLM 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 |
|---|---|---|---|---|---|---|---|
| SmolVLM-256M-Instruct | 256M | 0.6 GB | 0.3 GB | 0.1 GB | RTX 3070 | $0.088/hr | 0.70 GB |
| SmolVLM2-500M-Video-Instruct | 507M | 2.3 GB | 0.6 GB | 0.3 GB | RTX 3070 | $0.088/hr | 1.3 GB |
| SmolVLM2-2.2B-Instruct | 2.2B | 10.0 GB | 2.5 GB | 1.3 GB | V100 | $0.110/hr | not published for this architecture |
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 (4)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| SmolVLM-256M-Instruct | 256M | BF16 | 0.6 GB | RTX 3070 | $0.088/hr |
| SmolVLM-500M-Instruct | 507M | BF16 | 1.1 GB | RTX 3070 | $0.088/hr |
| SmolVLM2-500M-Video-Instruct | 507M | F32 | 2.3 GB | RTX 3070 | $0.088/hr |
| SmolVLM2-2.2B-Instruct | 2.2B | F32 | 10.0 GB | V100 | $0.110/hr |
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.