Text Generation
Transformers
PyTorch
Safetensors
mistral
custom_code
Eval Results (legacy)
text-generation-inference
Instructions to use d-matrix/Mistral-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use d-matrix/Mistral-7B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="d-matrix/Mistral-7B", trust_remote_code=True)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("d-matrix/Mistral-7B", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("d-matrix/Mistral-7B", trust_remote_code=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use d-matrix/Mistral-7B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "d-matrix/Mistral-7B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "d-matrix/Mistral-7B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/d-matrix/Mistral-7B
- SGLang
How to use d-matrix/Mistral-7B 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 "d-matrix/Mistral-7B" \ --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": "d-matrix/Mistral-7B", "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 "d-matrix/Mistral-7B" \ --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": "d-matrix/Mistral-7B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use d-matrix/Mistral-7B with Docker Model Runner:
docker model run hf.co/d-matrix/Mistral-7B
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -40,12 +40,12 @@ pip install -e .
|
|
| 40 |
from dmx.compressor.modeling import DmxModel
|
| 41 |
import lm_eval
|
| 42 |
|
| 43 |
-
model_args = "pretrained=
|
| 44 |
|
| 45 |
lm = lm_eval.api.registry.get_model("hf").create_from_arg_string(model_args, {"batch_size": 1})
|
| 46 |
|
| 47 |
# Transform the model with DMX
|
| 48 |
-
lm._model = DmxModel.from_torch(lm._model)
|
| 49 |
|
| 50 |
-
eval_results = lm_eval.evaluate(lm, lm_eval.tasks.get_task_dict([
|
| 51 |
```
|
|
|
|
| 40 |
from dmx.compressor.modeling import DmxModel
|
| 41 |
import lm_eval
|
| 42 |
|
| 43 |
+
model_args = "pretrained=d-matrix/Mistral,trust_remote_code=True"
|
| 44 |
|
| 45 |
lm = lm_eval.api.registry.get_model("hf").create_from_arg_string(model_args, {"batch_size": 1})
|
| 46 |
|
| 47 |
# Transform the model with DMX
|
| 48 |
+
lm._model = DmxModel.from_torch(lm._model)
|
| 49 |
|
| 50 |
+
eval_results = lm_eval.evaluate(lm, lm_eval.tasks.get_task_dict(["wikitext"])) # Assign desired task, i.e. "wikitext"
|
| 51 |
```
|