GGUF models by Lucy-in-the-Sky

265 GGUF models quantized by Lucy-in-the-Sky on Hugging Face. Sizes run from 494M to 46.7B parameters. The longest trained context is 1000K tokens.

Most downloadedNewestLargestLongest contextSmallest
MI
Dolphin Mistral 24B Venice Edition
Mistral · Lucy-in-the-Sky
Params23.6B
Context32K
Q4_K_M14.3 GB
GGUF
QW
Qwen2.5 14B Instruct Abliterated v2
Qwen · Lucy-in-the-Sky
Params14.8B
Context32K
Q4_K_M9.0 GB
Tool use
QW
Qwen3 4B Abliterated
Qwen · Lucy-in-the-Sky
Params4.02B
Context40K
Q4_K_M2.5 GB
ReasoningTool use
MI
Mixtral 8x7B Instruct v0.1
Mistral · Lucy-in-the-Sky
Params46.7B
Context32K
Q4_K_M28.4 GB
GGUF
QW
Qwen3 4B Abliterated
Qwen · Lucy-in-the-Sky
Params4.02B
Context40K
Smallest3.3 GB
ReasoningTool use
QW
Josiefied Qwen2.5 7B Instruct Abliterated v2
Qwen · Lucy-in-the-Sky
Params7.62B
Context32K
Q4_K_M4.7 GB
Tool use
MI
Dolphin Mistral 24B Venice Edition
Mistral · Lucy-in-the-Sky
Params23.6B
Context32K
Smallest19.3 GB
GGUF
NS
NSFW flash Q4 K M
Nsfw · Lucy-in-the-Sky
Params2.24B
Context4K
Q4_K_M1.4 GB
GGUF
QW
Qwen3 30B A3B
Qwen · Lucy-in-the-Sky
Params30.5B
Context40K
Q4_K_M18.6 GB
GGUF
QW
Qwen3 16B A3B
Qwen · Lucy-in-the-Sky
Params16B
Context40K
Q4_K_M9.8 GB
ReasoningTool use
QW
Qwen3 4B Abliterated
Qwen · Lucy-in-the-Sky
Params4.02B
Context40K
Smallest4.3 GB
ReasoningTool use
QW
Qwen2.5 Coder 14B Instruct
Qwen · Lucy-in-the-Sky
Params14.8B
Context32K
Q4_K_M9.0 GB
CodingTool use
QW
Qwen3 4B Abliterated
Qwen · Lucy-in-the-Sky
Params4.02B
Context40K
Smallest2.9 GB
ReasoningTool use
DE
Deepseek Moe 16b Base
DeepSeek · Lucy-in-the-Sky
Params16.4B
Context4K
Q4_K_M10.9 GB
GGUF
TA
UI TARS 1.5 7B
Tars · Lucy-in-the-Sky
Params7.62B
Context125K
Q4_K_M4.7 GB
Tool useVision
DE
Deepseek Moe 16b Chat
DeepSeek · Lucy-in-the-Sky
Params16.4B
Context4K
Smallest14.7 GB
GGUF
HE
NSFW 3B Q4 K M
Helvete · Lucy-in-the-Sky
Params2.8B
Context4K
Q4_K_M1.7 GB
GGUF
QW
Qwen2.5 Coder 32B Instruct
Qwen · Lucy-in-the-Sky
Params32.8B
Context32K
Q4_K_M19.9 GB
CodingTool use
LL
Dolphin 3.0 Llama 3.2 3B
Llama · Lucy-in-the-Sky
Params3.21B
Context128K
Smallest2.6 GB
CodingTool use
XO
XortronCriminalComputingConfig
Xortroncriminalcomputingconfig · Lucy-in-the-Sky
Params23.6B
Context32K
Q4_K_M14.3 GB
GGUF
QW
Qwen3 16B A3B
Qwen · Lucy-in-the-Sky
Params16B
Context40K
Smallest17.1 GB
ReasoningTool use
NS
NSFW flash Q8 0
Nsfw · Lucy-in-the-Sky
Params2.24B
Context4K
Smallest2.4 GB
GGUF
DE
Deepseek Moe 16b Chat
DeepSeek · Lucy-in-the-Sky
Params16.4B
Context4K
Smallest17.4 GB
GGUF
MI
Mistral Small 24B Instruct 2501
Mistral · Lucy-in-the-Sky
Params23.6B
Context32K
Q4_K_M14.3 GB
Reasoning
DE
Deepseek Coder 1.3b Instruct
DeepSeek · Lucy-in-the-Sky
Params1.35B
Context16K
Smallest1.4 GB
Coding
QW
Qwen2.5 14B Instruct Abliterated v2
Qwen · Lucy-in-the-Sky
Params14.8B
Context32K
Smallest15.7 GB
Tool use
DE
DeepSeek R1 Distill Qwen 14B
DeepSeek · Lucy-in-the-Sky
Params14.8B
Context128K
Q4_K_M9.0 GB
ReasoningTool use
DE
Deepseek Moe 16b Chat
DeepSeek · Lucy-in-the-Sky
Params16.4B
Context4K
Q4_K_M10.9 GB
GGUF
MA
Marco o1 Q8 0
Marco · Lucy-in-the-Sky
Params7.62B
Context32K
Smallest8.1 GB
GGUF
MI
Dolphin Mistral 24B Venice Edition
Mistral · Lucy-in-the-Sky
Params23.6B
Context32K
Smallest25.1 GB
GGUF

About Lucy-in-the-Sky on Hugging Face

Lucy-in-the-Sky has published 265 GGUF models that ggufmodels.com tracks, across families such as Qwen, DeepSeek and Mistral. The most downloaded are Dolphin Mistral 24B Venice Edition, Qwen2.5 14B Instruct Abliterated v2 and Qwen3 4B Abliterated.

The quants Lucy-in-the-Sky publishes most often are Q8_0 (82), Q4_K_M (71), Q6_K (54) and Q2_K (46). Their Q4_K_M files range from 0.9 to 28.4 GB. Parameter counts run from 494M to 46.7B, and the longest trained context window is 1000K tokens.

Frequently asked questions

What GGUF models has Lucy-in-the-Sky published?

265 models are listed here, mainly from the Qwen, DeepSeek and Mistral families. The most downloaded is Dolphin Mistral 24B Venice Edition.

Which quants does Lucy-in-the-Sky publish?

The most common are Q8_0 (82 models), Q4_K_M (71 models), Q6_K (54 models) and Q2_K (46 models).

Keep browsing