Model comparison
SmolVLM2-2.2B-Instruct vs embeddinggemma-300m
These two publish different quantization sets; the table below has the exact sizes.
From the file· summed bytes, KV per layer
Side by side
| SmolVLM2-2.2B-Instruct | embeddinggemma-300m | |
|---|---|---|
| Parameters | 2.2B | 303M |
| Architecture | llama | gemma-embedding |
| Layers | 24 | 24 |
| Native context | 8,192 | 2,048 |
| Mixture of experts | no | no |
| Quantizations published | 40 | 10 |
| Smallest quantization | 0.42 GiB | 0.26 GiB |
| Q4_K_M | 1.04 GiB | — |
| Licence | apache-2.0 | — |
KV cache by context
the term that decides long-context viability
| Context | SmolVLM2-2.2B-Instruct | embeddinggemma-300m | Ratio |
|---|---|---|---|
| 4,096 | — | 0.04 GiB | — |
| 8,192 | — | 0.05 GiB | — |
| 16,384 | — | 0.08 GiB | — |
| 32,768 | — | 0.14 GiB | — |
| 65,536 | — | 0.27 GiB | — |
| 131,072 | — | 0.52 GiB | — |