KaLM Embedding Gemma3 12B 2511 vs Qwen 3 Embedding 4B
KaLM Embedding Gemma3 12B 251172.3
Qwen 3 Embedding 4B69.4
| Benchmark | KaLM Embedding Gemma3 12B 2511 | Qwen 3 Embedding 4B |
|---|---|---|
| MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages) | 72.3 | 69.4 |
Specs
| Spec | KaLM Embedding Gemma3 12B 2511 | Qwen 3 Embedding 4B |
|---|---|---|
| Price | — | !$0.02 |
| Context | 33K | 33K |
| Weights | Open | Open |
| Released | Nov 2025 | Jun 2025 |
| Dimensions | 3,840 | 2,560 |
| Max input | 131K tokens | 33K tokens |
| Memory (4-bit) | 7.1 GB | 2.4 GB |
| Licence | tencent-kalm-embedding-community | Apache-2.0 |
| Commercial use | Restricted | Allowed |
What breaks when you switch
KaLM Embedding Gemma3 12B 2511 → Qwen 3 Embedding 4B
Re-embed your whole corpus›
Vectors from different models can't be mixed; dimensions change from 3,840 to 2,560.
Max input: 131K tokens → 33K tokens
Qwen 3 Embedding 4B → KaLM Embedding Gemma3 12B 2511
Re-embed your whole corpus›
Vectors from different models can't be mixed; dimensions change from 2,560 to 3,840.