Instructions to use moondream/moondream-2b-2025-04-14-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use moondream/moondream-2b-2025-04-14-4bit with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="moondream/moondream-2b-2025-04-14-4bit",
	filename="moondream2-mmproj-f16.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use moondream/moondream-2b-2025-04-14-4bit with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf moondream/moondream-2b-2025-04-14-4bit:F16
# Run inference directly in the terminal:
llama-cli -hf moondream/moondream-2b-2025-04-14-4bit:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf moondream/moondream-2b-2025-04-14-4bit:F16
# Run inference directly in the terminal:
llama-cli -hf moondream/moondream-2b-2025-04-14-4bit:F16

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf moondream/moondream-2b-2025-04-14-4bit:F16
# Run inference directly in the terminal:
./llama-cli -hf moondream/moondream-2b-2025-04-14-4bit:F16

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf moondream/moondream-2b-2025-04-14-4bit:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf moondream/moondream-2b-2025-04-14-4bit:F16

Use Docker

docker model run hf.co/moondream/moondream-2b-2025-04-14-4bit:F16

LM Studio
Jan

vLLM

How to use moondream/moondream-2b-2025-04-14-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "moondream/moondream-2b-2025-04-14-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "moondream/moondream-2b-2025-04-14-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/moondream/moondream-2b-2025-04-14-4bit:F16

Ollama
How to use moondream/moondream-2b-2025-04-14-4bit with Ollama:
```
ollama run hf.co/moondream/moondream-2b-2025-04-14-4bit:F16
```

Unsloth Studio

How to use moondream/moondream-2b-2025-04-14-4bit with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for moondream/moondream-2b-2025-04-14-4bit to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for moondream/moondream-2b-2025-04-14-4bit to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for moondream/moondream-2b-2025-04-14-4bit to start chatting

Atomic Chat new
Docker Model Runner
How to use moondream/moondream-2b-2025-04-14-4bit with Docker Model Runner:
```
docker model run hf.co/moondream/moondream-2b-2025-04-14-4bit:F16
```

Lemonade

How to use moondream/moondream-2b-2025-04-14-4bit with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull moondream/moondream-2b-2025-04-14-4bit:F16

Run and chat with the model

lemonade run user.moondream-2b-2025-04-14-4bit-F16

List all available models

lemonade list

snowclipsed commited on Jun 9, 2025

Commit

1db79dd

1 Parent(s): 8cb4b5e

add group size as a parameter

Browse files

Files changed (3) hide show

layers.py +6 -5
moondream.py +6 -4
text.py +4 -4

layers.py CHANGED Viewed

@@ -36,26 +36,27 @@ class QuantizedLinear(nn.Module):
         self,
         in_features: int,
         out_features: int,
         dtype: torch.dtype = torch.uint8,
     ):
-        # TODO: Take group_size as an input instead of hardcoding it here.
         super().__init__()
         self.in_features = in_features
         self.out_features = out_features
         self.weight = nn.ParameterDict(
             {
                 "packed": nn.Parameter(
                     torch.empty(
-                        out_features * in_features // (128 * 2), 128, dtype=dtype
                     ),
                     requires_grad=False,
                 ),
                 "scale": nn.Parameter(
-                    torch.empty(out_features * in_features // 128, 1),
                     requires_grad=False,
                 ),
                 "zero_point": nn.Parameter(
-                    torch.empty(out_features * in_features // 128, 1),
                     requires_grad=False,
                 ),
             }
@@ -86,7 +87,7 @@ class QuantizedLinear(nn.Module):
         )
         del self.weight, self.bias
-        quantize_(self, int4_weight_only(group_size=128))
         self.unpacked = True
         torch.cuda.empty_cache()

         self,
         in_features: int,
         out_features: int,
+        group_size: int = 128,
         dtype: torch.dtype = torch.uint8,
     ):
         super().__init__()
         self.in_features = in_features
         self.out_features = out_features
+        self.group_size = group_size
         self.weight = nn.ParameterDict(
             {
                 "packed": nn.Parameter(
                     torch.empty(
+                        out_features * in_features // (group_size * 2), group_size, dtype=dtype
                     ),
                     requires_grad=False,
                 ),
                 "scale": nn.Parameter(
+                    torch.empty(out_features * in_features // group_size, 1),
                     requires_grad=False,
                 ),
                 "zero_point": nn.Parameter(
+                    torch.empty(out_features * in_features // group_size, 1),
                     requires_grad=False,
                 ),
             }
         )
         del self.weight, self.bias
+        quantize_(self, int4_weight_only(group_size=self.group_size))
         self.unpacked = True
         torch.cuda.empty_cache()

moondream.py CHANGED Viewed

@@ -80,31 +80,33 @@ class MoondreamModel(nn.Module):
         self.region = nn.ModuleDict(
             {
                 "coord_encoder": QuantizedLinear(
-                    config.region.coord_feat_dim, config.region.dim, dtype=dtype
                 ),
                 "coord_decoder": nn.ModuleDict(
                     {
                         "fc1": QuantizedLinear(
-                            config.region.dim, config.region.inner_dim, dtype=dtype
                         ),
                         "fc2": QuantizedLinear(
                             config.region.inner_dim,
                             config.region.coord_out_dim,
                             dtype=dtype,
                         ),
                     }
                 ),
                 "size_encoder": QuantizedLinear(
-                    config.region.size_feat_dim, config.region.dim, dtype=dtype
                 ),
                 "size_decoder": nn.ModuleDict(
                     {
                         "fc1": QuantizedLinear(
-                            config.region.dim, config.region.inner_dim, dtype=dtype
                         ),
                         "fc2": QuantizedLinear(
                             config.region.inner_dim,
                             config.region.size_out_dim,
                             dtype=dtype,
                         ),
                     }

         self.region = nn.ModuleDict(
             {
                 "coord_encoder": QuantizedLinear(
+                    config.region.coord_feat_dim, config.region.dim, group_size=config.text.group_size, dtype=dtype
                 ),
                 "coord_decoder": nn.ModuleDict(
                     {
                         "fc1": QuantizedLinear(
+                            config.region.dim, config.region.inner_dim, group_size=config.text.group_size, dtype=dtype
                         ),
                         "fc2": QuantizedLinear(
                             config.region.inner_dim,
                             config.region.coord_out_dim,
+                            group_size=config.text.group_size,
                             dtype=dtype,
                         ),
                     }
                 ),
                 "size_encoder": QuantizedLinear(
+                    config.region.size_feat_dim, config.region.dim, group_size=config.text.group_size, dtype=dtype
                 ),
                 "size_decoder": nn.ModuleDict(
                     {
                         "fc1": QuantizedLinear(
+                            config.region.dim, config.region.inner_dim, group_size=config.text.group_size, dtype=dtype
                         ),
                         "fc2": QuantizedLinear(
                             config.region.inner_dim,
                             config.region.size_out_dim,
+                            group_size=config.text.group_size,
                             dtype=dtype,
                         ),
                     }

text.py CHANGED Viewed

@@ -164,19 +164,19 @@ def build_text_model(config: TextConfig, dtype: torch.dtype = torch.float16) ->
                             "ln": nn.LayerNorm(config.dim, dtype=dtype),
                             "attn": nn.ModuleDict(
                                 {
-                                    "qkv": QuantizedLinear(config.dim, qkv_dim, dtype=dtype),
                                     "proj": QuantizedLinear(
-                                        config.dim, config.dim, dtype=dtype
                                     ),
                                 }
                             ),
                             "mlp": nn.ModuleDict(
                                 {
                                     "fc1": QuantizedLinear(
-                                        config.dim, config.ff_dim, dtype=dtype
                                     ),
                                     "fc2": QuantizedLinear(
-                                        config.ff_dim, config.dim, dtype=dtype
                                     ),
                                 }
                             ),

                             "ln": nn.LayerNorm(config.dim, dtype=dtype),
                             "attn": nn.ModuleDict(
                                 {
+                                    "qkv": QuantizedLinear(config.dim, qkv_dim, group_size=config.text.group_size, dtype=dtype),
                                     "proj": QuantizedLinear(
+                                        config.dim, config.dim, group_size=config.text.group_size, dtype=dtype
                                     ),
                                 }
                             ),
                             "mlp": nn.ModuleDict(
                                 {
                                     "fc1": QuantizedLinear(
+                                        config.dim, config.ff_dim, group_size=config.text.group_size, dtype=dtype
                                     ),
                                     "fc2": QuantizedLinear(
+                                        config.ff_dim, config.dim, group_size=config.text.group_size, dtype=dtype
                                     ),
                                 }
                             ),