KaLM Embedding Multilingual Mini Instruct v1 vs KaLM Embedding Multilingual Mini v1
No benchmarks in this slice.
Specs
| Spec | KaLM Embedding Multilingual Mini Instruct v1 | KaLM Embedding Multilingual Mini v1 |
|---|---|---|
| Context | 1K | 1K |
| Weights | Open | Open |
| Released | Oct 2024 | Aug 2024 |
| Dimensions | 896 | 896 |
| Max input | 33K tokens | 1K tokens |
| Memory (4-bit) | 0.3 GB | 0.3 GB |
| Licence | MIT | MIT |
| Commercial use | Allowed | Allowed |
What breaks when you switch
KaLM Embedding Multilingual Mini Instruct v1 → KaLM Embedding Multilingual Mini v1
Re-embed your whole corpus›
Vectors from different models can't be mixed.
Max input: 33K tokens → 1K tokens
KaLM Embedding Multilingual Mini v1 → KaLM Embedding Multilingual Mini Instruct v1
Re-embed your whole corpus›
Vectors from different models can't be mixed.