Fabert measured on 9 of 261 Embeddings benchmarks9
BenchmarkByScoreRank
No benchmark matches
price not verifiedprice
50.6MTEB Persian v2
Oct 2024released
768Dimensions
1K tokensMax input
PersianLanguages
Alternatives?
ModelScoreCostPriceWhat changes
ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
ModelScoreCostPriceWhat changes
cF2LLM v2 8BunstatedOpen+57%——Memory (4-bit): 0.1 GB → 4.5 GBcF2LLM v2 4BunstatedOpen+55%——Memory (4-bit): 0.1 GB → 2.4 GBcF2LLM v2 1.7BunstatedOpen+52%——Memory (4-bit): 0.1 GB → 1 GBcF2LLM v2 0.6BunstatedOpen+47%——Memory (4-bit): 0.1 GB → 0.4 GBRun it yourself
estimateFits on✓ 8 GB (phone, laptop) · full quality✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
Licencenot verified
Benchmarks
in Embeddings & RetrievalFAQ
How much does Fabert cost?›
There is no per-unit API price: price not verified.
How good is Fabert, and what is it best at?›
#61 of 149 in Embeddings.
Is Fabert better than TookaBERT Base?›
On MTEB Persian v2 · Mean over tasks · Persian, Fabert scores 46.4 vs 46.5 — about the same.
When was Fabert released? Is there a newer version?›
Released 2024-10-07. It is the current version in its line.
Is Fabert open source?›
Yes, the weights are open (not specified licence).
Can I run Fabert locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Fabert?›
Fabert is an open-weights embeddings & retrieval model from sbunlp, released Oct 2024.
What are the best alternatives to Fabert?›
EmbeddingGemma 300M (better), F2LLM v2 8B (open weights).