GGUF models by richardyoung

35 GGUF models quantized by richardyoung on Hugging Face. Sizes run from 596M to 480B parameters. The longest trained context is 256K tokens.

Most downloadedNewestLargestLongest contextSmallest
QW
Qwen3 14B Abliterated Conservative
Qwen · richardyoung
Params14.8B
Context40K
Q4_K_M9.0 GB
GGUF
QW
Qwen2.5 7B Instruct Abliterated
Qwen · richardyoung
Params7.62B
Context32K
Q4_K_M4.7 GB
GGUF
QW
Qwen3 8b Abliterated
Qwen · richardyoung
Params8.19B
Context40K
Q4_K_M5.0 GB
GGUF
DE
DeepSeek R1 0528 Qwen3 8B heretic
DeepSeek · richardyoung
Params8.2B
Context128K
Q4_K_M5.0 GB
Reasoning
LI
LightOnOCR 3 4B
Lightonocr · richardyoung
Params4.2B
Context256K
Q4_K_M2.7 GB
Vision
AG
Agents A1 4B heretic
Agents · richardyoung
Params4.2B
Context256K
Q4_K_M2.7 GB
Vision
FR
FrogNano 4B 2609 heretic
Frognano · richardyoung
Params4.2B
Context256K
Q4_K_M2.7 GB
Vision
LF
LFM2.5 24B A2B
Lfm · richardyoung
Params24B
Context128K
Q4_K_M14.5 GB
ReasoningTool use
PH
Phi 4 mini instruct heretic
Phi · richardyoung
Params3.84B
Context128K
Q4_K_M2.5 GB
GGUF
LF
LFM2.5 1.2B Instruct heretic
Lfm · richardyoung
Params1.17B
Context125K
Q4_K_M0.7 GB
GGUF
WE
WeVisDoc 4B
Wevisdoc · richardyoung
Params4.41B
Context256K
Q4_K_M2.7 GB
Tool useVision
LI
LightOnOCR 3 0.8B
Lightonocr · richardyoung
Params752M
Context256K
Q4_K_M0.5 GB
ReasoningTool useVision
GR
granite 4.2 8b heretic
Granite · richardyoung
Params8.79B
Context128K
Q4_K_M5.3 GB
ReasoningTool use
OL
Olmo 3 7B Think heretic
OLMo · richardyoung
Params7.3B
Context64K
Q4_K_M4.5 GB
Reasoning
XI
Xiaomi OCR 0
Xiaomi · richardyoung
Params773M
Context256K
Q4_K_M0.5 GB
ReasoningTool useVision
GR
granite 4.2 3b heretic
Granite · richardyoung
Params3.66B
Context128K
Q4_K_M2.2 GB
ReasoningTool use
FR
FrogNano 4B 2609
Frognano · richardyoung
Params4.33B
Context256K
Q4_K_M2.8 GB
ReasoningTool useVision
MU
Muse Glimmer 30B heretic
Muse · richardyoung
Params28B
Context128K
Q4_K_M16.9 GB
ReasoningVision
LI
LightOnOCR 3 1B
Lightonocr · richardyoung
Params596M
Context16K
Q4_K_M0.4 GB
Vision
KI
Kimi VL A3B Thinking 2506
Kimi · richardyoung
Params16B
Context128K
Q4_K_M10.5 GB
ReasoningVision
WE
WeVisDoc 2B
Wevisdoc · richardyoung
Params2.03B
Context256K
Q4_K_M1.3 GB
Tool useVision
DO
dots.mocr
Dots · richardyoung
Params1.78B
Context128K
Q4_K_M1.1 GB
Vision
HE
Hemmingway 1 heretic
Hemmingway · richardyoung
Params27.3B
Context256K
Q4_K_M16.8 GB
GGUF
BE
Bev 9B inverted
Bev · richardyoung
Params8.95B
Context256K
Q4_K_M5.6 GB
GGUF
MI
MiniCPM V 4.6
MiniCPM · richardyoung
Params752M
Context256K
Q4_K_M0.5 GB
ReasoningTool useVision
UN
Unlimited OCR
Unlimited · richardyoung
Params2.93B
Context32K
Q4_K_M2.0 GB
GGUF
OL
Olmo 3 7B RLZero Math
OLMo · richardyoung
Params7.3B
Context64K
Q4_K_M4.5 GB
Reasoning
OL
olmOCR 2 7B 1025
OLMo · richardyoung
Params7.62B
Context125K
Smallest8.1 GB
Vision
QW
Qwen3.5 4B SOM Heretic
Qwen · richardyoung
Params4.21B
Context256K
Q4_K_M2.7 GB
ReasoningTool use
QW
Qwen2.5 Coder 7B Instruct Heretic1X General
Qwen · richardyoung
Params7.62B
Context32K
Q4_K_M4.7 GB
CodingTool use

About richardyoung on Hugging Face

richardyoung has published 35 GGUF models that ggufmodels.com tracks, across families such as Qwen, OLMo and Lightonocr. The most downloaded are Qwen3 14B Abliterated Conservative, Qwen2.5 7B Instruct Abliterated and Qwen3 8b Abliterated.

The quants richardyoung publishes most often are Q4_K_M (30), Q8_0 (28), Q5_K_M (23) and Q6_K (21). Their Q4_K_M files range from 0.4 to 19.9 GB. Parameter counts run from 596M to 480B, and the longest trained context window is 256K tokens.

Frequently asked questions

What GGUF models has richardyoung published?

35 models are listed here, mainly from the Qwen, OLMo and Lightonocr families. The most downloaded is Qwen3 14B Abliterated Conservative.

Which quants does richardyoung publish?

The most common are Q4_K_M (30 models), Q8_0 (28 models), Q5_K_M (23 models) and Q6_K (21 models).

Keep browsing