GGUF models: mpl-2.0 license

34 GGUF models listed under the mpl-2.0 license on Hugging Face. Always read the license text before commercial use. Sizes run from 38M to 14.8B parameters. The longest trained context is 1024K tokens.

Most downloadedNewestLargestLongest contextSmallest
XT
XTTS v2 GPT
Xtts · onicubag
Params434M
Context—
Q4_K_M0.3 GB
GGUF
LI
Lightning 1.7B
Lightning · bartowski
Params1.72B
Context40K
Q4_K_M1.1 GB
Tool use
QW
gaiasky qwen 3.5
Qwen · Langurmonkey
Params4.21B
Context256K
Q4_K_M2.7 GB
CodingReasoningTool use
LI
Linden 4B FP32
Linden · mradermacher
Params4.21B
Context256K
Q4_K_M2.7 GB
GGUF
HO
HomePhi4_4B_FP16
Homephi · mradermacher
Params3.84B
Context128K
Q4_K_M2.5 GB
Tool use
LI
Linden 4B FP32
Linden · mradermacher
Params4.21B
Context256K
Q4_K_M2.7 GB
GGUF
HO
HomePhi4_4B_FP16
Homephi · mradermacher
Params3.84B
Context128K
Q4_K_M2.5 GB
Tool use
FO
Foresight 135M
Foresight · mradermacher
Params135M
Context8K
Q4_K_M0.1 GB
GGUF
PY
pyGPT 50M
Pygpt · mradermacher
Params77M
Context1K
Q4_K_M0.1 GB
Coding
SO
Solomon 0.5B FP16 Checkpoint 1000
Solomon · mradermacher
Params494M
Context32K
Q4_K_M0.4 GB
ReasoningTool use
EV
Eve 4b
Eve · TitleOS
Params4.41B
Context256K
Smallest3.2 GB
CodingTool use
SO
Solomon Nano 350M FP32
Solomon · mradermacher
Params494M
Context32K
Q4_K_M0.4 GB
ReasoningTool use
EV
Eve 4b FP16
Eve · TitleOS
Params4.41B
Context256K
Smallest8.8 GB
CodingTool use
QW
qwen3 1.7b tool
Qwen · Igriscodes
Params1.72B
Context40K
Q4_K_M1.1 GB
ReasoningTool use
LO
LokiHA 2B
Lokiha · TitleOS
Params1.88B
Context256K
Smallest1.4 GB
ReasoningTool use
EV
Eve 4b FP16
Eve · mradermacher
Params4.02B
Context256K
Q4_K_M2.5 GB
CodingTool use
FR
Friday 14B FP16
Friday · mradermacher
Params14.8B
Context128K
Q4_K_M9.0 GB
Tool use
HO
HomePhi4 4B Q4 K M
Homephi · TitleOS
Params3.84B
Context128K
Q4_K_M2.5 GB
Tool use
QW
qwen3 4b tool
Qwen · Igriscodes
Params4.02B
Context40K
Q4_K_M2.5 GB
ReasoningTool use
QW
buleyean qwen2.5 0.5b
Qwen · forkjoin-ai
Params494M
Context32K
Q4_K_M0.4 GB
Tool use
PY
pyGPT 50M
Pygpt · mradermacher
Params77M
Context1K
Q4_K_M0.1 GB
Coding
KI
Kizuna 7B 20240509
Kizuna · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF
SP
Spark 270M FP16
Spark · TitleOS
Params268M
Context32K
Q4_K_M0.3 GB
GGUF
SP
Spark 270M FP16
Spark · TitleOS
Params268M
Context32K
Smallest0.3 GB
GGUF
AD
ADBait 1B FP16
Adbait · mradermacher
Params1.46B
Context1024K
Q4_K_M0.9 GB
GGUF
QW
Linden 4B FP32
Qwen · TitleOS
Params4.21B
Context256K
Q4_K_M2.7 GB
GGUF
EV
Eve 4b FP16
Eve · mradermacher
Params4.02B
Context256K
Q4_K_M2.5 GB
CodingTool use
SM
buleyean smollm2 360m
SmolLM · forkjoin-ai
Params362M
Context8K
Q4_K_M0.3 GB
GGUF
HO
HomePhi4 4B Q8 0
Homephi · TitleOS
Params3.84B
Context128K
Smallest4.1 GB
Tool use
QW
Lightning 1.7B
Qwen · TitleOS
Params1.72B
Context40K
Smallest1.8 GB
Tool use

About the mpl-2.0 license

The license text is in the model repository on Hugging Face. Licenses can differ between a quantized file and the original model, so check both.

34 GGUF models are listed under mpl-2.0 on Hugging Face. The most downloaded are XTTS v2 GPT, Lightning 1.7B and gaiasky qwen 3.5. This is information, not legal advice: always read the license in the repository before you use a model commercially.

Frequently asked questions

Can I use mpl-2.0 GGUF models commercially?

The license text is in the model repository on Hugging Face. Licenses can differ between a quantized file and the original model, so check both. Check the license text in the repository and the license of the base model before you build a product on it.

Keep browsing