GGUF models by gabriellarson

87 GGUF models quantized by gabriellarson on Hugging Face. Sizes run from 354M to 1030B parameters. The longest trained context is 1024K tokens.

Most downloadedNewestLargestLongest contextSmallest
MO
Moonlight 16B A3B Instruct
Moonlight · gabriellarson
Params16B
Context8K
Q4_K_M10.5 GB
GGUF
HE
Hermes 4 14B
Hermes · gabriellarson
Params14.8B
Context40K
Q4_K_M9.0 GB
ReasoningRoleplayTool use
PH
Phi Mini MoE
Phi · gabriellarson
Params7.65B
Context4K
Q4_K_M5.0 GB
GGUF
UI
UIGEN T3.1 8B Full Demo No Reasoning
Uigen · gabriellarson
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
MI
Devstral Small 2507
Mistral · gabriellarson
Params23.6B
Context128K
Q4_K_M14.3 GB
CodingTool useVision
MI
Mistral Small 3.2 24B Instruct 2506
Mistral · gabriellarson
Params23.6B
Context128K
Q4_K_M14.3 GB
Vision
KI
Kimi Dev 72B
Kimi · gabriellarson
Params72.7B
Context128K
Q4_K_M47.4 GB
CodingTool use
NE
OpenReasoning Nemotron 1.5B
Nemotron · gabriellarson
Params1.54B
Context128K
Q4_K_M1.0 GB
CodingReasoning
MI
Mamba Codestral 7B v0.1
Mistral · gabriellarson
Params7.29B
Context1024K
Smallest1.5 GB
Coding
HU
Hunyuan A13B Instruct
Hunyuan · gabriellarson
Params80.4B
Context256K
Q4_K_M48.8 GB
ReasoningTool use
QW
Qwen3 4b
Qwen · gabriellarson
Params4.02B
Context256K
Q4_K_M2.5 GB
ReasoningTool use
NF
NFT 32B
Nft · gabriellarson
Params32.8B
Context128K
Q4_K_M19.9 GB
ReasoningTool use
NE
OpenCodeReasoning Nemotron 1.1 14B
Nemotron · gabriellarson
Params14.8B
Context64K
Q4_K_M9.0 GB
CodingReasoningTool use
SE
Seed OSS 36B Instruct
Seed · gabriellarson
Params36.2B
Context512K
Q4_K_M21.8 GB
ReasoningTool use
FA
Falcon H1 34B Instruct
Falcon · gabriellarson
Params33.6B
Context256K
Q4_K_M20.3 GB
Tool use
JA
Jan nano 128k
Jan · gabriellarson
Params4.02B
Context128K
Q4_K_M2.5 GB
ReasoningTool use
NE
OpenReasoning Nemotron 14B
Nemotron · gabriellarson
Params14.8B
Context128K
Q4_K_M9.0 GB
CodingReasoning
NE
OpenCodeReasoning Nemotron 1.1 7B
Nemotron · gabriellarson
Params7.62B
Context32K
Q4_K_M4.7 GB
CodingReasoningTool use
LF
LFM2 VL 450M
Lfm · gabriellarson
Params354M
Context125K
Q4_K_M0.2 GB
Vision
SM
SmallThinker 4BA0.6B Instruct
Smallthinker · gabriellarson
Params4.27B
Context32K
Q4_K_M2.6 GB
Tool use
NE
NextCoder 7B
Nextcoder · gabriellarson
Params7.62B
Context32K
Q4_K_M4.7 GB
CodingTool use
FO
Foundation Sec 8B Instruct
Foundation · gabriellarson
Params8.03B
Context128K
Q4_K_M4.9 GB
GGUF
LF
LFM2 8B A1B
Lfm · gabriellarson
Params8.34B
Context125K
Q4_K_M5.0 GB
Tool use
RE
Reka Flash 3.1
Reka · gabriellarson
Params20.9B
Context96K
Q4_K_M13.6 GB
GGUF
SM
SmallThinker 21BA3B Instruct
Smallthinker · gabriellarson
Params21.5B
Context16K
Q4_K_M13.1 GB
Tool use
QW
OpenBuddy R1 0528 Distill Qwen3 32B Preview6 QAT 200K
Qwen · gabriellarson
Params32.8B
Context195K
Q4_K_M19.8 GB
ReasoningTool use
SK
Skywork SWE 32B
Skywork · gabriellarson
Params32.8B
Context32K
Q4_K_M19.9 GB
Tool use
LF
LFM2 VL 1.6B
Lfm · gabriellarson
Params1.17B
Context125K
Q4_K_M0.7 GB
Vision
QW
Qwen3 4B Instruct 2507
Qwen · gabriellarson
Params4.02B
Context256K
Q4_K_M2.5 GB
ReasoningTool use
KI
Kimina Prover Distill 8B
Kimi · gabriellarson
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use

About gabriellarson on Hugging Face

gabriellarson has published 87 GGUF models that ggufmodels.com tracks, across families such as Nemotron, Qwen and Kimi. The most downloaded are Moonlight 16B A3B Instruct, Hermes 4 14B and Phi Mini MoE.

The quants gabriellarson publishes most often are Q8_0 (81), Q6_K (80), Q4_0 (80) and Q5_K_M (79). Their Q4_K_M files range from 0.2 to 240.3 GB. Parameter counts run from 354M to 1030B, and the longest trained context window is 1024K tokens.

Frequently asked questions

What GGUF models has gabriellarson published?

87 models are listed here, mainly from the Nemotron, Qwen and Kimi families. The most downloaded is Moonlight 16B A3B Instruct.

Which quants does gabriellarson publish?

The most common are Q8_0 (81 models), Q6_K (80 models), Q4_0 (80 models) and Q5_K_M (79 models).

Keep browsing