GGUF models by jacobcarajo

55 GGUF models quantized by jacobcarajo on Hugging Face. Sizes run from 494M to 34.7B parameters. The longest trained context is 256K tokens.

Most downloadedNewestLargestLongest contextSmallest
DE
DeepSeek R1 Distill Qwen 14B
DeepSeek · jacobcarajo
Params14.8B
Context128K
Smallest10.5 GB
ReasoningTool use
QW
Qwen3 Coder 30B A3B Instruct
Qwen · jacobcarajo
Params30.5B
Context256K
Smallest21.7 GB
CodingTool use
LL
dolphin 2.9.4 llama3.1 8b Q5 K M
Llama · jacobcarajo
Params8.03B
Context128K
Smallest5.7 GB
Coding
QW
Qwen2.5 32B Instruct
Qwen · jacobcarajo
Params32.8B
Context32K
Smallest23.3 GB
Tool use
QW
Qwen2.5 Coder 7B Instruct
Qwen · jacobcarajo
Params7.62B
Context32K
Smallest5.4 GB
CodingTool use
QW
Qwen2.5 7B Instruct
Qwen · jacobcarajo
Params7.62B
Context32K
Smallest5.4 GB
Tool use
MI
Dolphin3.0 Mistral 24B
Mistral · jacobcarajo
Params23.6B
Context32K
Smallest16.8 GB
CodingTool use
MI
Dolphin Mistral 24B Venice Edition
Mistral · jacobcarajo
Params23.6B
Context128K
Smallest16.8 GB
GGUF
QW
Qwen2.5 14B Instruct
Qwen · jacobcarajo
Params14.8B
Context32K
Smallest10.5 GB
Tool use
CY
CyberStrike OffSec 35B
Cyberstrike · jacobcarajo
Params34.7B
Context256K
Smallest24.7 GB
Tool use
QW
Huihui Qwen3.5 27B Abliterated
Qwen · jacobcarajo
Params27.3B
Context256K
Smallest19.5 GB
ReasoningTool useVision
PH
phi 4 Q5 K M
Phi · jacobcarajo
Params14.7B
Context16K
Smallest10.6 GB
Coding
LL
Phind CodeLlama 34B v2
Llama · jacobcarajo
Params33.7B
Context16K
Smallest23.8 GB
Coding
MI
Mistral 7B Instruct v0.3
Mistral · jacobcarajo
Params7.25B
Context32K
Smallest5.1 GB
Tool use
LL
Llama 3.1 8B Instruct
Llama · jacobcarajo
Params8.03B
Context128K
Smallest5.7 GB
Tool use
QW
Qwen3 30B A3B Instruct 2507
Qwen · jacobcarajo
Params30.5B
Context256K
Smallest21.7 GB
ReasoningTool use
MI
Mistral Small 3.1 24B Instruct 2503
Mistral · jacobcarajo
Params23.6B
Context128K
Smallest16.8 GB
Vision
DE
DeepSeek R1 Distill Qwen 32B
DeepSeek · jacobcarajo
Params32.8B
Context128K
Smallest23.3 GB
ReasoningTool use
QW
Qwen3.5 9B
Qwen · jacobcarajo
Params9.2B
Context256K
Smallest6.6 GB
ReasoningTool useVision
QW
Qwen3.5 27B
Qwen · jacobcarajo
Params27.3B
Context256K
Smallest19.5 GB
ReasoningTool useVision
DE
DeepSeek R1 Distill Qwen 32B Abliterated
DeepSeek · jacobcarajo
Params32.8B
Context128K
Smallest23.3 GB
ReasoningTool use
GE
Gemma 2 9b It
Gemma · jacobcarajo
Params9.24B
Context8K
Smallest6.6 GB
GGUF
DE
DeepSeek Coder V2 Lite Instruct
DeepSeek · jacobcarajo
Params15.7B
Context160K
Smallest11.9 GB
Coding
WI
WizardCoder Python 34B V1.0 Q5 K M
Wizardcoder · jacobcarajo
Params33.7B
Context16K
Smallest23.8 GB
Coding
MI
dolphin 2.9.3 mistral 7B 32k Q5 K M
Mistral · jacobcarajo
Params7.25B
Context32K
Smallest5.1 GB
CodingTool use
GP
Huihui Gpt Oss 20b BF16 Abliterated
gpt-oss · jacobcarajo
Params20.9B
Context128K
Smallest16.9 GB
GGUF
CO
Codegemma 7b
Codegemma · jacobcarajo
Params8.54B
Context8K
Smallest6.1 GB
GGUF
QW
Qwen2 Math 7B Instruct
Qwen · jacobcarajo
Params7.62B
Context4K
Smallest5.4 GB
GGUF
LL
Meta Llama 3.1 8B
Llama · jacobcarajo
Params8.03B
Context128K
Smallest8.5 GB
GGUF
MI
Dolphin3.0 R1 Mistral 24B
Mistral · jacobcarajo
Params23.6B
Context32K
Smallest16.8 GB
CodingReasoningTool use

About jacobcarajo on Hugging Face

jacobcarajo has published 55 GGUF models that ggufmodels.com tracks, across families such as Qwen, Mistral and Llama. The most downloaded are DeepSeek R1 Distill Qwen 14B, Qwen3 Coder 30B A3B Instruct and dolphin 2.9.4 llama3.1 8b Q5 K M.

The quants jacobcarajo publishes most often are Q5_K_M (51), Q4_K_M (2), Q8_0 (1) and BF16 (1). Their Q4_K_M files range from 0.4 to 4.9 GB. Parameter counts run from 494M to 34.7B, and the longest trained context window is 256K tokens.

Frequently asked questions

What GGUF models has jacobcarajo published?

55 models are listed here, mainly from the Qwen, Mistral and Llama families. The most downloaded is DeepSeek R1 Distill Qwen 14B.

Which quants does jacobcarajo publish?

The most common are Q5_K_M (51 models), Q4_K_M (2 models), Q8_0 (1 models) and BF16 (1 models).

Keep browsing