meta llama3 (#210)

frontend/core/blocks/meta-llama/Dockerfile

@@ -0,0 +1,29 @@
+# Use the CUDA-enabled base image with Python 3.9
+FROM nvidia/cuda:11.8.0-cudnn8-runtime-ubuntu22.04
+
+# Set the working directory
+WORKDIR /app
+
+# Install Python, Git, and necessary dependencies
+RUN apt-get update && apt-get install -y \
+    python3-pip \
+    python3-dev \
+    git \
+    && ln -s /usr/bin/python3 /usr/bin/python \
+    && rm -rf /var/lib/apt/lists/*
+
+
+# Install PyTorch with CUDA support
+RUN pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
+
+# Copy the requirements file
+COPY requirements.txt .
+
+# Install other Python dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy the Python script
+COPY computations.py .
+
+
+

frontend/core/blocks/meta-llama/computations.py

@@ -0,0 +1,23 @@
+import transformers
+import torch
+from huggingface_hub import login
+
+
+def compute(hf_token, model_id, prompt):
+    """
+    meta-llama
+    """
+    # Login with your Hugging Face token and save it to the Git credentials helper
+    login(token=hf_token, add_to_git_credential=True)
+    pipeline = transformers.pipeline(
+        "text-generation", model=model_id, model_kwargs={"torch_dtype": torch.bfloat16}, device="cuda"
+    )
+    output = pipeline(prompt)
+
+    return {"llama": output}
+
+
+def test():
+    """Test the compute function."""
+
+    print("Running test")

frontend/core/blocks/meta-llama/requirements.txt

@@ -0,0 +1,3 @@
+transformers==4.43.1
+numpy==1.26.4
+accelerate==0.33.0

frontend/core/blocks/meta-llama/specs.json

@@ -0,0 +1,75 @@
+{
+  "information": {
+    "id": "meta-llama",
+    "name": "Llama",
+    "description": "meta-llama",
+    "system_versions": [
+      "0.1"
+    ],
+    "block_version": "block version number",
+    "block_source": "core/blocks/meta-llama",
+    "block_type": "compute"
+  },
+  "inputs": {
+    "hf_token": {
+      "type": "Any",
+      "connections": [
+
+      ]
+    },
+    "model_id": {
+      "type": "Any",
+      "connections": [
+
+      ]
+    },
+    "prompt": {
+      "type": "Any",
+      "connections": [
+
+      ]
+    }
+  },
+  "outputs": {
+    "llama": {
+      "type": "Any",
+      "connections": []
+    }
+  },
+  "action": {
+    "container": {
+      "image": "meta-llama-gpu1",
+      "version": "latest",
+      "command_line": [
+        "python",
+        "-u",
+        "entrypoint.py"
+      ]
+    }
+  },
+  "views": {
+    "node": {
+      "active": "True or False",
+      "title_bar": {
+        "background_color": "#6b2be0"
+      },
+      "preview": {},
+      "html": "",
+      "pos_x": "1045",
+      "pos_y": "197",
+      "pos_z": "999",
+      "behavior": "modal",
+      "order": {
+        "input": [
+          "hf_token",
+          "model_id",
+          "prompt"
+        ],
+        "output": [
+          "llama"
+        ]
+      }
+    }
+  },
+  "events": {}
+}