Model comparison
jina-embeddings-v5-text-nano vs nomic-embed-text-v1
At Q4_K_M, nomic-embed-text-v1 is the smaller download — 84,106,624 bytes against 156,966,976.
From the file· summed bytes, KV per layer
Side by side
| jina-embeddings-v5-text-nano | nomic-embed-text-v1 | |
|---|---|---|
| Parameters | 212M | 137M |
| Architecture | eurobert | nomic-bert |
| Layers | 12 | 12 |
| Native context | 8,192 | 8,192 |
| Mixture of experts | no | no |
| Quantizations published | 42 | 14 |
| Smallest quantization | 0.09 GiB | 0.05 GiB |
| Q4_K_M | 0.15 GiB | 0.08 GiB |
| Licence | cc-by-nc-4.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | jina-embeddings-v5-text-nano | nomic-embed-text-v1 | Ratio |
|---|---|---|---|
| 4,096 | 0.14 GiB | 0.14 GiB | 1.00× |
| 8,192 | 0.28 GiB | 0.28 GiB | 1.00× |
| 16,384 | 0.56 GiB | 0.56 GiB | 1.00× |
| 32,768 | 1.13 GiB | 1.13 GiB | 1.00× |
| 65,536 | 2.25 GiB | 2.25 GiB | 1.00× |
| 131,072 | 4.50 GiB | 4.50 GiB | 1.00× |