Instructions to use FastFlowLM/Gemma3-4B-NPU2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use FastFlowLM/Gemma3-4B-NPU2 with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("FastFlowLM/Gemma3-4B-NPU2", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Unsloth Studio
How to use FastFlowLM/Gemma3-4B-NPU2 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 FastFlowLM/Gemma3-4B-NPU2 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 FastFlowLM/Gemma3-4B-NPU2 to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for FastFlowLM/Gemma3-4B-NPU2 to start chatting
Load model with FastModel
pip install unsloth from unsloth import FastModel model, tokenizer = FastModel.from_pretrained( model_name="FastFlowLM/Gemma3-4B-NPU2", max_seq_length=2048, )
Update config.json
Browse files- config.json +1 -1
config.json
CHANGED
|
@@ -32,7 +32,7 @@
|
|
| 32 |
"addr_l_begin_mha": 53760,
|
| 33 |
"addr_l_end_mha": 25088,
|
| 34 |
"addr_kk": 45056,
|
| 35 |
-
"flm_version": "0.9.
|
| 36 |
"vision_model_weight": "vision_weight.q4nx",
|
| 37 |
"vision_mm_engine_xclbin_name": "vision_mm.xclbin",
|
| 38 |
"vision_mha_engine_xclbin_name":"vision_attn.xclbin",
|
|
|
|
| 32 |
"addr_l_begin_mha": 53760,
|
| 33 |
"addr_l_end_mha": 25088,
|
| 34 |
"addr_kk": 45056,
|
| 35 |
+
"flm_version": "0.9.11",
|
| 36 |
"vision_model_weight": "vision_weight.q4nx",
|
| 37 |
"vision_mm_engine_xclbin_name": "vision_mm.xclbin",
|
| 38 |
"vision_mha_engine_xclbin_name":"vision_attn.xclbin",
|