Model comparison
mxbai-embed-xsmall-v1 vs nomic-embed-text-v1
These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: mxbai-embed-xsmall-v1's KV cache at 32K is 4.0× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| mxbai-embed-xsmall-v1 | nomic-embed-text-v1 | |
|---|---|---|
| Parameters | 24M | 137M |
| Architecture | bert | nomic-bert |
| Layers | 6 | 12 |
| Native context | 4,096 | 8,192 |
| Mixture of experts | no | no |
| Quantizations published | 4 | 14 |
| Smallest quantization | 0.03 GiB | 0.05 GiB |
| Q4_K_M | — | 0.08 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | mxbai-embed-xsmall-v1 | nomic-embed-text-v1 | Ratio |
|---|---|---|---|
| 4,096 | 0.04 GiB | 0.14 GiB | 4.00× |
| 8,192 | 0.07 GiB | 0.28 GiB | 4.00× |
| 16,384 | 0.14 GiB | 0.56 GiB | 4.00× |
| 32,768 | 0.28 GiB | 1.13 GiB | 4.00× |
| 65,536 | 0.56 GiB | 2.25 GiB | 4.00× |
| 131,072 | 1.13 GiB | 4.50 GiB | 4.00× |