First commit

Dockerfile

@@ -1,13 +1,9 @@
 FROM python:3.10
 
 WORKDIR /app
-
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
 COPY . .
 
-# Make the script executable
-RUN chmod +x run.sh
-
-CMD ["bash", "run.sh"]
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,10 +1,16 @@
----
-title: Deepseek 7b
-emoji: 🐢
-colorFrom: red
-colorTo: green
-sdk: docker
-pinned: false
----
+# DeepSeek API Server
+This is a FastAPI-based text generation API using the DeepSeek-R1 7B model.
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+## Installation
+1. Clone this repository.
+2. Install dependencies: `pip install -r requirements.txt`
+3. Run the server: `uvicorn app:app --host 0.0.0.0 --port 7860`
+
+## Usage
+Send a POST request to `/generate` with a JSON payload:
+
+```json
+{
+  "prompt": "What is AI?",
+  "max_tokens": 100
+}

app.py

@@ -0,0 +1,34 @@
+import torch
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
+
+# Model setup
+MODEL_NAME = "deepseek-ai/deepseek-llm-7b-base"
+DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+
+# Load model and tokenizer
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+model = AutoModelForCausalLM.from_pretrained(
+    MODEL_NAME, torch_dtype=torch.bfloat16, device_map="auto"
+)
+model.generation_config = GenerationConfig.from_pretrained(MODEL_NAME)
+model.generation_config.pad_token_id = model.generation_config.eos_token_id
+
+# FastAPI app
+app = FastAPI()
+
+# Request payload
+class TextGenerationRequest(BaseModel):
+    prompt: str
+    max_tokens: int = 100
+
+@app.post("/generate")
+async def generate_text(request: TextGenerationRequest):
+    try:
+        inputs = tokenizer(request.prompt, return_tensors="pt").to(DEVICE)
+        outputs = model.generate(**inputs, max_new_tokens=request.max_tokens)
+        result = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        return {"generated_text": result}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))

app/api.py

@@ -1,14 +0,0 @@
-from fastapi import FastAPI
-from pydantic import BaseModel
-from app.model import model
-
-app = FastAPI()
-
-class PromptRequest(BaseModel):
-    prompt: str
-    max_length: int = 256
-
-@app.post("/generate")
-async def generate_text(request: PromptRequest):
-    result = model.generate(request.prompt, request.max_length)
-    return {"response": result}

app/config.py

@@ -1,5 +0,0 @@
-import os
-
-MODEL_NAME = "deepseek-ai/deepseek-llm-7b-base"
-HF_TOKEN = os.getenv("HF_TOKEN")  # Set this in the Hugging Face environment
-DEVICE = "cuda"  # Change to "cpu" if no GPU

app/model.py

@@ -1,19 +0,0 @@
-import os
-from transformers import AutoModelForCausalLM, AutoTokenizer
-import torch
-from app.config import MODEL_NAME, DEVICE
-
-class DeepSeekModel:
-    def __init__(self):
-        self.tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, token=os.getenv("HF_TOKEN"))
-        self.model = AutoModelForCausalLM.from_pretrained(
-            MODEL_NAME, torch_dtype=torch.float16, device_map="auto"
-        )
-
-    def generate(self, prompt, max_length=256):
-        inputs = self.tokenizer(prompt, return_tensors="pt").to(DEVICE)
-        with torch.no_grad():
-            outputs = self.model.generate(**inputs, max_length=max_length)
-        return self.tokenizer.decode(outputs[0], skip_special_tokens=True)
-
-model = DeepSeekModel()

gitignore

@@ -0,0 +1,4 @@
+__pycache__/
+*.pyc
+*.pyo
+*.log

requirements.txt

@@ -1,5 +1,4 @@
-fastapi
-uvicorn
 torch
 transformers
-accelerate
+fastapi
+uvicorn

run.sh

@@ -1,2 +0,0 @@
-#!/bin/bash
-uvicorn app.api:app --host 0.0.0.0 --port 8000 --workers 1