Instructions to use vikhyatk/moondream1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use vikhyatk/moondream1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="vikhyatk/moondream1", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("vikhyatk/moondream1", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use vikhyatk/moondream1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "vikhyatk/moondream1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "vikhyatk/moondream1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/vikhyatk/moondream1
- SGLang
How to use vikhyatk/moondream1 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "vikhyatk/moondream1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "vikhyatk/moondream1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "vikhyatk/moondream1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "vikhyatk/moondream1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use vikhyatk/moondream1 with Docker Model Runner:
docker model run hf.co/vikhyatk/moondream1
proposed patch for the Tensor size mismatch error
#15
by claverAI - opened
- modeling_phi.py +1 -0
modeling_phi.py
CHANGED
|
@@ -315,6 +315,7 @@ class CrossAttention(nn.Module):
|
|
| 315 |
dtype=scores.dtype,
|
| 316 |
device=scores.device,
|
| 317 |
)
|
|
|
|
| 318 |
padding_mask.masked_fill_(key_padding_mask, 0.0)
|
| 319 |
scores = scores + rearrange(padding_mask, "b s -> b 1 1 s")
|
| 320 |
|
|
|
|
| 315 |
dtype=scores.dtype,
|
| 316 |
device=scores.device,
|
| 317 |
)
|
| 318 |
+
key_padding_mask = key_padding_mask[:, :seqlen_k]
|
| 319 |
padding_mask.masked_fill_(key_padding_mask, 0.0)
|
| 320 |
scores = scores + rearrange(padding_mask, "b s -> b 1 1 s")
|
| 321 |
|