GGUF models by sinjab

55 GGUF models quantized by sinjab on Hugging Face. Sizes run from 4M to 7.57B parameters. The longest trained context is 40K tokens.

Most downloadedNewestLargestLongest contextSmallest
MA
ms marco MiniLM L6 v2 Q8 0
Marco · sinjab
Params23M
Context512
Smallest0.0 GB
CodingEmbedding
QW
Qwen3 Reranker 4B
Qwen · sinjab
Params4.02B
Context40K
Smallest8.0 GB
EmbeddingReasoningTool use
BG
bge reranker base Q8 0
BGE · sinjab
Params278M
Context512
Smallest0.3 GB
Embedding
BG
bge reranker v2 m3 Q8 0
BGE · sinjab
Params568M
Context8K
Smallest0.6 GB
Embedding
BG
bge reranker v2 m3 F16
BGE · sinjab
Params568M
Context8K
Smallest1.2 GB
Embedding
JI
Jina Reranker v2 Base Multilingual
Jina · sinjab
Params278M
Context1K
Smallest0.6 GB
Embedding
QW
Qwen3 Reranker 4B
Qwen · sinjab
Params4.02B
Context40K
Smallest4.3 GB
EmbeddingReasoningTool use
QW
Qwen3 Reranker 4B
Qwen · sinjab
Params4.02B
Context40K
Q4_K_M2.5 GB
EmbeddingReasoningTool use
MA
ms marco MiniLM L6 v2 F16
Marco · sinjab
Params23M
Context512
Smallest0.0 GB
CodingEmbedding
BG
bge reranker v2 m3 Q4 K M
BGE · sinjab
Params568M
Context8K
Q4_K_M0.4 GB
Embedding
BG
bge reranker base Q4 K M
BGE · sinjab
Params278M
Context512
Q4_K_M0.2 GB
Embedding
JI
Jina Reranker v2 Base Multilingual
Jina · sinjab
Params278M
Context1K
Q4_K_M0.2 GB
Embedding
BG
bge reranker base F16
BGE · sinjab
Params278M
Context512
Smallest0.6 GB
Embedding
JI
jina reranker v1 turbo en F16
Jina · sinjab
Params38M
Context8K
Smallest0.1 GB
Embedding
JI
Jina Reranker v2 Base Multilingual
Jina · sinjab
Params278M
Context1K
Smallest0.3 GB
Embedding
QW
Qwen3 Reranker 8B
Qwen · sinjab
Params7.57B
Context40K
Q4_K_M4.7 GB
EmbeddingReasoningTool use
QW
Qwen3 Reranker 0.6B
Qwen · sinjab
Params596M
Context40K
Smallest1.2 GB
EmbeddingReasoningTool use
MA
ms marco TinyBERT L2 F16
Marco · sinjab
Params4M
Context512
Smallest0.0 GB
CodingEmbedding
MA
ms marco MiniLM L6 v2 Q4 K M
Marco · sinjab
Params23M
Context512
Q4_K_M0.0 GB
CodingEmbedding
JI
jina reranker v1 turbo en Q4 K M
Jina · sinjab
Params38M
Context8K
Q4_K_M0.0 GB
Embedding
MX
Mxbai Rerank Base v2
Mxbai · sinjab
Params494M
Context32K
Smallest0.5 GB
EmbeddingTool use
JI
jina reranker v1 turbo en Q8 0
Jina · sinjab
Params38M
Context8K
Smallest0.0 GB
Embedding
MA
ms marco MiniLM L12 v2 Q4 K M
Marco · sinjab
Params33M
Context512
Q4_K_M0.0 GB
CodingEmbedding
QW
Qwen3 Reranker 0.6B
Qwen · sinjab
Params596M
Context40K
Smallest0.6 GB
EmbeddingReasoningTool use
MX
Mxbai Rerank Base v2
Mxbai · sinjab
Params494M
Context32K
Q4_K_M0.4 GB
EmbeddingTool use
MX
Mxbai Rerank Large v2
Mxbai · sinjab
Params1.54B
Context32K
Q4_K_M1.0 GB
EmbeddingTool use
MA
ms marco MiniLM L4 v2 Q4 K M
Marco · sinjab
Params19M
Context512
Q4_K_M0.0 GB
CodingEmbedding
QW
Qwen3 Reranker 8B
Qwen · sinjab
Params7.57B
Context40K
Smallest15.1 GB
EmbeddingReasoningTool use
BG
bge reranker large Q8 0
BGE · sinjab
Params560M
Context512
Smallest0.6 GB
Embedding
QW
Qwen3 Reranker 0.6B
Qwen · sinjab
Params596M
Context40K
Q4_K_M0.4 GB
EmbeddingReasoningTool use

About sinjab on Hugging Face

sinjab has published 55 GGUF models that ggufmodels.com tracks, across families such as Marco, Qwen and Jina. The most downloaded are ms marco MiniLM L6 v2 Q8 0, Qwen3 Reranker 4B and bge reranker base Q8 0.

The quants sinjab publishes most often are F16 (19), Q8_0 (18) and Q4_K_M (18). Their Q4_K_M files range from 0.0 to 4.7 GB. Parameter counts run from 4M to 7.57B, and the longest trained context window is 40K tokens.

Frequently asked questions

What GGUF models has sinjab published?

55 models are listed here, mainly from the Marco, Qwen and Jina families. The most downloaded is ms marco MiniLM L6 v2 Q8 0.

Which quants does sinjab publish?

The most common are F16 (19 models), Q8_0 (18 models) and Q4_K_M (18 models).

Keep browsing