GGUF models by build-small-hackathon

41 GGUF models quantized by build-small-hackathon on Hugging Face. Sizes run from 11M to 31.6B parameters. The longest trained context is 1024K tokens.

Most downloadedNewestLargestLongest contextSmallest
VE
velvet rope lenore lora
Velvet · build-small-hackathon
Params33M
Context—
Smallest0.1 GB
GGUF
HA
halide vision
Halide · build-small-hackathon
Params752M
Context256K
Q4_K_M0.5 GB
ReasoningTool useVision
VE
velvet rope vivienne lora
Velvet · build-small-hackathon
Params33M
Context—
Smallest0.1 GB
GGUF
QW
proofkit distilled qwen0.5b
Qwen · build-small-hackathon
Params494M
Context32K
Q4_K_M0.4 GB
Tool use
GE
agenda parser medium
Gemma · build-small-hackathon
Params25.2B
Context256K
Q4_K_M16.8 GB
ReasoningTool use
GE
storygrove gemma 3 4b
Gemma · build-small-hackathon
Params3.88B
Context128K
Q4_K_M2.5 GB
GGUF
VE
velvet rope aurelia lora
Velvet · build-small-hackathon
Params33M
Context—
Smallest0.1 GB
GGUF
QW
legacyscribe Qwen3.5 9B.Q4 K M
Qwen · build-small-hackathon
Params9.2B
Context256K
Q4_K_M5.8 GB
ReasoningTool use
VE
velvet rope marlowe lora
Velvet · build-small-hackathon
Params33M
Context—
Smallest0.1 GB
GGUF
VE
velvet rope crispin lora
Velvet · build-small-hackathon
Params33M
Context—
Smallest0.1 GB
GGUF
UN
untranslator chronicler
Untranslator · build-small-hackathon
Params3.09B
Context32K
Smallest1.9 GB
Tool use

Keep browsing