Instructions to use InstaDeepAI/ChatNT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use InstaDeepAI/ChatNT with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="InstaDeepAI/ChatNT", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("InstaDeepAI/ChatNT", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use InstaDeepAI/ChatNT with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "InstaDeepAI/ChatNT" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "InstaDeepAI/ChatNT", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/InstaDeepAI/ChatNT
- SGLang
How to use InstaDeepAI/ChatNT 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 "InstaDeepAI/ChatNT" \ --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": "InstaDeepAI/ChatNT", "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 "InstaDeepAI/ChatNT" \ --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": "InstaDeepAI/ChatNT", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use InstaDeepAI/ChatNT with Docker Model Runner:
docker model run hf.co/InstaDeepAI/ChatNT
Update chatNT.py
Browse files
chatNT.py
CHANGED
|
@@ -1763,8 +1763,8 @@ class TorchMultiModalPerceiverResampler(nn.Module):
|
|
| 1763 |
concat_input_1 = torch.cat([xf_1, x], dim=1)
|
| 1764 |
concat_input_2 = torch.cat([xf_2, x], dim=1)
|
| 1765 |
|
| 1766 |
-
|
| 1767 |
-
|
| 1768 |
|
| 1769 |
output = layer(
|
| 1770 |
x=x,
|
|
@@ -1774,7 +1774,7 @@ class TorchMultiModalPerceiverResampler(nn.Module):
|
|
| 1774 |
attention_mask_2=attention_mask_2,
|
| 1775 |
)
|
| 1776 |
x = output["embeddings"]
|
| 1777 |
-
|
| 1778 |
|
| 1779 |
return x, outs
|
| 1780 |
|
|
|
|
| 1763 |
concat_input_1 = torch.cat([xf_1, x], dim=1)
|
| 1764 |
concat_input_2 = torch.cat([xf_2, x], dim=1)
|
| 1765 |
|
| 1766 |
+
outs[f"PERCEIVER_RESAMPLER_concat_input_1_{layer_idx}"] = concat_input_1.clone()
|
| 1767 |
+
outs[f"PERCEIVER_RESAMPLER_concat_input_2_{layer_idx}"] = concat_input_2.clone()
|
| 1768 |
|
| 1769 |
output = layer(
|
| 1770 |
x=x,
|
|
|
|
| 1774 |
attention_mask_2=attention_mask_2,
|
| 1775 |
)
|
| 1776 |
x = output["embeddings"]
|
| 1777 |
+
outs[f"PERCEIVER_RESAMPLER_attention_embeddings_{layer_idx}"] = output["embeddings"].clone()
|
| 1778 |
|
| 1779 |
return x, outs
|
| 1780 |
|