All tasks / Comparisons
F2LLM-v2-14B vs KaLM-Embedding-Gemma3-12B-2511
Compared only on public benchmark results, across the 1 task where both are listed: KaLM-Embedding-Gemma3-12B-2511 is a tier higher in 1 of 1 shared task.
F2LLM-v2-14B
Open-weights Apache-2.0 multilingual text embedding models from codefuse-ai, from 80M to 14B parameters.
- Vendor
- Ant Group (codefuse-ai)
- Weights
- Open: run it yourself
- Released
- Mar 2026
- GPU memory to run locally
- ~28 GB
- License
- Apache-2.0
- Last verified
- 19 Oct 2026
KaLM-Embedding-Gemma3-12B-2511
Tencent's open-weights 12B multilingual text embedding model built on Gemma 3.
- Vendor
- Tencent
- Weights
- Open: run it yourself
- Released
- Nov 2025
- GPU memory to run locally
- ~24 GB
- License
- Tencent KaLM Embedding Community License
- Last verified
- 19 Oct 2026
Task by task
| Task | F2LLM-v2-14B | KaLM-Embedding-Gemma3-12B-2511 | Higher tier |
|---|---|---|---|
| Embeddings & retrieval Models | A68.7% on MTEB(Multilingual, v2) · Local | S72.3% on MTEB(Multilingual, v2) · Local | KaLM-Embedding-Gemma3-12B-2511 |
A tier spans a range of scores, and small gaps on a leaderboard are often within its margin of error, so only a different tier counts as ahead. How we rank.
Same benchmark, both results
| Benchmark | F2LLM-v2-14B | KaLM-Embedding-Gemma3-12B-2511 | Snapshot |
|---|---|---|---|
| MTEB(Multilingual, v2) Mean (Task): average main score over 131 multilingual tasks (bitext mining, classification, clustering, retrieval, reranking, STS, instruction reranking), higher is better | 68.7% | 72.3% | 19 Oct 2026 |