Hugging Face's logo

KaLM-Embedding
/
KaLM-Reranker-V1-Nano-R2

Text Ranking
sentence-transformers
Safetensors
Transformers
multilingual
t5gemma2
text2text-generation
reranker
encoder-decoder
FBNL
matryoshka
retrieval
RAG
Model card Files
xet
 Community
Yuki131 commited on
Commit
0d20167
·
verified ·
1 Parent(s): 9f4bf5c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -5
README.md CHANGED
@@ -79,11 +79,12 @@ All results below use the same default compression ratio, `r=4`. KaLM-Reranker-V
79
  | Multilingual ranking | Limited | **Substantially improved across MIRACL's 18 languages** |
80
 
81
  ## Model family
82
- | Model | Activated parameters | Layers (encoder + decoder) | Sequence length | Document token dimension | MEP support | Instruction aware |
83
- |:--|--:|--:|--:|--:|:--:|:--:|
84
- | [KaLM-Reranker-V1-Nano-R2](https://huggingface.co/KaLM-Embedding/KaLM-Reranker-V1-Nano-R2) | 0.27B | 18 + 18 | 128K | 640 | 1x-128x | Yes |
85
- | [KaLM-Reranker-V1-Small-R2](https://huggingface.co/KaLM-Embedding/KaLM-Reranker-V1-Small-R2) | 1B | 26 + 26 | 128K | 1152 | 1x-128x | Yes |
86
- | [KaLM-Reranker-V1-Large-R2](https://huggingface.co/KaLM-Embedding/KaLM-Reranker-V1-Large-R2) | 4B | 34 + 34 | 128K | 2560 | 1x-128x | Yes |
 
87
 
88
  The reported sizes are **activated parameters**. Nano, Small, and Large are initialized from the T5Gemma2 270M-270M, 1B-1B, and 4B-4B encoder-decoder families, respectively.
89
 
 
79
  | Multilingual ranking | Limited | **Substantially improved across MIRACL's 18 languages** |
80
 
81
  ## Model family
82
+
83
+ | Models | Activated Params. | Non-Embedding Params. | Embedding Params. | #Layers (encoder + decoder) | Sequence Length | Document Token Dim. | MEP Support | Instruction Aware | Test-Time Compute |
84
+ | ------------------------------------------------------------------------------------------------- | -----------------: | --------------------: | ----------------: | --------------------------: | --------------: | ------------------: | ----------: | ----------------: | ----------------: |
85
+ | [KaLM-Reranker-V1-Nano-R2](https://huggingface.co/KaLM-Embedding/KaLM-Reranker-V1-Nano-R2) | 0.27B | 100M | 168M | 18+18 | 128K | 640 | 1x-128x | Yes | Yes |
86
+ | [KaLM-Reranker-V1-Small-R2](https://huggingface.co/KaLM-Embedding/KaLM-Reranker-V1-Small-R2) | 1B | 698M | 302M | 26+26 | 128K | 1152 | 1x-128x | Yes | Yes |
87
+ | [KaLM-Reranker-V1-Large-R2](https://huggingface.co/KaLM-Embedding/KaLM-Reranker-V1-Large-R2) | 4B | 3209M | 675M | 34+34 | 128K | 2560 | 1x-128x | Yes | Yes |
88
 
89
  The reported sizes are **activated parameters**. Nano, Small, and Large are initialized from the T5Gemma2 270M-270M, 1B-1B, and 4B-4B encoder-decoder families, respectively.
90