Instructions to use hbx/Mistral-Interact with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use hbx/Mistral-Interact with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="hbx/Mistral-Interact", device_map="auto")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("hbx/Mistral-Interact") model = AutoModelForCausalLM.from_pretrained("hbx/Mistral-Interact", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use hbx/Mistral-Interact with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "hbx/Mistral-Interact" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "hbx/Mistral-Interact", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/hbx/Mistral-Interact
- SGLang
How to use hbx/Mistral-Interact 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 "hbx/Mistral-Interact" \ --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": "hbx/Mistral-Interact", "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 "hbx/Mistral-Interact" \ --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": "hbx/Mistral-Interact", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use hbx/Mistral-Interact with Docker Model Runner:
docker model run hf.co/hbx/Mistral-Interact
Update README.md
Browse files
README.md
CHANGED
|
@@ -8,7 +8,13 @@ datasets:
|
|
| 8 |
|
| 9 |
# Model Card for Mistral-Interact
|
| 10 |
|
| 11 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
- **Better understanding of user judgments:** Among all the open-source models, Mistral-Interact is the best at predicting task vagueness and missing details that users regard as necessary.
|
| 13 |
- **Comprehensive summarization of user intentions:** Mistral-Interact is effective in making an explicit and comprehensive summary based on detailed user intentions.
|
| 14 |
- **Enhanced model-user interaction experience:** Mistral-Interact inquires about missing details in vague tasks more reasonably and friendly than other open-source models, thus promoting a clearer understanding of the user’s implicit intentions.
|
|
|
|
| 8 |
|
| 9 |
# Model Card for Mistral-Interact
|
| 10 |
|
| 11 |
+
- Base IN3: https://huggingface.co/datasets/hbx/IN3
|
| 12 |
+
- IN3-interaction: https://huggingface.co/datasets/hbx/IN3-interaction
|
| 13 |
+
- Paper: https://arxiv.org/abs/2402.09205
|
| 14 |
+
- Model: https://huggingface.co/hbx/Mistral-Interact
|
| 15 |
+
- Repo: https://github.com/HBX-hbx/Mistral-Interact
|
| 16 |
+
|
| 17 |
+
Using the [constructed interaction data](https://huggingface.co/datasets/hbx/IN3-interaction), we adapt Mistral-7B into Mistral-Interact, a powerful and robust variant of Mistral, capable of judging the vagueness of user instruction, actively querying for missing details with suggestions, and explicitly summarizing the detailed and clear user intentions. It has the following features:
|
| 18 |
- **Better understanding of user judgments:** Among all the open-source models, Mistral-Interact is the best at predicting task vagueness and missing details that users regard as necessary.
|
| 19 |
- **Comprehensive summarization of user intentions:** Mistral-Interact is effective in making an explicit and comprehensive summary based on detailed user intentions.
|
| 20 |
- **Enhanced model-user interaction experience:** Mistral-Interact inquires about missing details in vague tasks more reasonably and friendly than other open-source models, thus promoting a clearer understanding of the user’s implicit intentions.
|