| SPECIFICATIONS | |||||
| Max dimension | 4K | 2K | 4K | 1K | |
| Max tokens | 33K | 33K | 4K | 8K | |
| Parameters | 8B | 3.8B | 7B | 559M | |
| License | Apache 2.0 | Apache 2.0 | Apache 2.0 | CC BY-NC 4.0 | |
| EMBEDDING TYPES | |||||
| Types |
Dense
|
DenseLate
|
Dense
|
Dense
|
|
| MATRYOSHKA DIMENSIONS | |||||
| Available sizes |
3264128256512102420484096
|
12825651210242048
|
— |
32641282565121024
|
|
| INPUT MODALITIES | |||||
| Text | ✓ | ✓ | ✓ | ✓ | |
| Image | — | ✓ | — | — | |
| — | ✓ | — | — | ||
| OUTPUT TYPES | |||||
| Single vector | ✓ | ✓ | ✓ | ✓ | |
| Multi vector | — | ✓ | — | — | |
| LANGUAGE SUPPORT | |||||
| Languages | 🌍 100+ | 🌍 29+ | 🌍 Multi | 🌍 89+ | |
| PRICING | |||||
| Per 1M tokens | — | $0.050 | — | $0.020 | |
| BENCHMARKS | |||||
| CMTEB CHINESE | 73.84 | — | — | — | |
| COIR | — | 71.59 | — | — | |
| JINA VDR | — | 84.11 | — | — | |
| LONGEMBED | — | 67.11 | — | 55.66 | |
| MMTEB | — | 66.49 | — | 58.58 | |
| MTEB AVG | — | — | 68.20 | — | |
| MTEB EN | — | 55.97 | — | 54.33 | |
| MTEB EN V2 | 75.22 | — | — | — | |
| MTEB MULTILINGUAL | 70.58 | — | — | — | |
| MTEB RETRIEVAL | — | — | 60.20 | — | |
| VIDORE | — | 90.17 | — | — | |