F2LLM v2 1.7B vs F2LLM v2 4B
F2LLM v2 1.7B65.2
F2LLM v2 4B67.1
| Benchmark | F2LLM v2 1.7B | F2LLM v2 4B |
|---|---|---|
| MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages) | 65.2 | 67.1 |
Specs
| Spec | F2LLM v2 1.7B | F2LLM v2 4B |
|---|---|---|
| Context | 41K | 41K |
| Weights | Open | Open |
| Released | Mar 2026 | Mar 2026 |
| Dimensions | 2,048 | 2,560 |
| Max input | 41K tokens | 41K tokens |
| Languages | en, es, Estonian +80 | en, es, Estonian +80 |
| Memory (4-bit) | 1 GB | 2.4 GB |
| Licence | Apache-2.0 | Apache-2.0 |
| Commercial use | Allowed | Allowed |
What breaks when you switch
F2LLM v2 1.7B → F2LLM v2 4B
Re-embed your whole corpus›
Vectors from different models can't be mixed; dimensions change from 2,048 to 2,560.
Memory (4-bit): 1 GB → 2.4 GB
F2LLM v2 4B → F2LLM v2 1.7B
Re-embed your whole corpus›
Vectors from different models can't be mixed; dimensions change from 2,560 to 2,048.