Tarka Embedding 150M v1 vs Tarka Embedding 350M v1
No benchmarks in this slice.
Specs
| Spec | Tarka Embedding 150M v1 | Tarka Embedding 350M v1 |
|---|---|---|
| Context | 2K | 128K |
| Weights | Open | Open |
| Released | Nov 2025 | Nov 2025 |
| Dimensions | — | 1,024 |
| Max input | 2K tokens | 128K tokens |
| Languages | en, es, fr +5 | en, es, fr +5 |
| Memory (4-bit) | 0.1 GB | 0.2 GB |
| Licence | Gemma terms | lfm1.0 |
| Commercial use | Restricted | Restricted |
What breaks when you switch
Tarka Embedding 150M v1 → Tarka Embedding 350M v1
Re-embed your whole corpus›
Vectors from different models can't be mixed.
Memory (4-bit): 0.1 GB → 0.2 GB
Tarka Embedding 350M v1 → Tarka Embedding 150M v1
Re-embed your whole corpus›
Vectors from different models can't be mixed.