Instructions to use darkps/darkit-v2.5-transformers with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use darkps/darkit-v2.5-transformers with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="darkps/darkit-v2.5-transformers")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import darkit-v2.5
model = darkit-v2.5.from_pretrained("darkps/darkit-v2.5-transformers", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use darkps/darkit-v2.5-transformers with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "darkps/darkit-v2.5-transformers"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "darkps/darkit-v2.5-transformers",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/darkps/darkit-v2.5-transformers

SGLang

How to use darkps/darkit-v2.5-transformers 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 "darkps/darkit-v2.5-transformers" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "darkps/darkit-v2.5-transformers",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "darkps/darkit-v2.5-transformers" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "darkps/darkit-v2.5-transformers",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use darkps/darkit-v2.5-transformers with Docker Model Runner:
```
docker model run hf.co/darkps/darkit-v2.5-transformers
```

darkps commited on 4 days ago

Commit

93e74a9

verified ·

1 Parent(s): 4de68f2

Upload notebook.ipynb

Browse files

Files changed (1) hide show

notebook.ipynb +25 -25

notebook.ipynb CHANGED Viewed

@@ -19,29 +19,29 @@
         "from transformers import AutoModelForCausalLM, AutoTokenizer\n",
         "import torch\n",
         "\n",
-        "REPO_ID = \"darkps/darkit-v2.5-transformers\"\n",
         "\n",
-        "api = HfApi()\n",
         "\n",
-        "files = api.list_repo_files(REPO_ID)\n",
         "\n",
-        "print(\"Repository files:\")\n",
-        "for f in files:\n",
-        "    print(f)\n",
         "\n",
-        "tokenizer = AutoTokenizer.from_pretrained(\n",
-        "    REPO_ID,\n",
         "    trust_remote_code=True\n",
         ")\n",
         "\n",
-        "model = AutoModelForCausalLM.from_pretrained(\n",
-        "    REPO_ID,\n",
         "    torch_dtype=torch.float16,\n",
         "    device_map=\"auto\",\n",
         "    trust_remote_code=True\n",
         ")\n",
         "\n",
-        "model.eval()\n"
       ]
     },
     {
@@ -50,43 +50,43 @@
       "metadata": {},
       "outputs": [],
       "source": [
-        "PROMPT = \"Hi how are you?\"\n",
         "\n",
-        "messages = [\n",
         "    {\n",
         "        \"role\": \"user\",\n",
-        "        \"content\": PROMPT\n",
         "    }\n",
         "]\n",
         "\n",
-        "text = tokenizer.apply_chat_template(\n",
-        "    messages,\n",
         "    tokenize=False,\n",
         "    add_generation_prompt=True\n",
         ")\n",
         "\n",
-        "inputs = tokenizer(\n",
-        "    text,\n",
         "    return_tensors=\"pt\"\n",
-        ").to(model.device)\n",
         "\n",
         "with torch.no_grad():\n",
-        "    outputs = model.generate(\n",
-        "        **inputs,\n",
         "        max_new_tokens=128,\n",
         "        temperature=0.7,\n",
         "        top_p=0.8,\n",
         "        top_k=20,\n",
         "        do_sample=True,\n",
-        "        eos_token_id=tokenizer.eos_token_id\n",
         "    )\n",
         "\n",
-        "response = tokenizer.decode(\n",
-        "    outputs[0][inputs.input_ids.shape[-1]:],\n",
         "    skip_special_tokens=True\n",
         ")\n",
         "\n",
-        "print(response)\n"
       ]
     }
   ],

         "from transformers import AutoModelForCausalLM, AutoTokenizer\n",
         "import torch\n",
         "\n",
+        "ИДЕНТИФИКАТОР_РЕПО = \"darkps/darkit-v2.5-transformers\"\n",
         "\n",
+        "АПИ = HfApi()\n",
         "\n",
+        "СПИСОК_ФАЙЛОВ = АПИ.list_repo_files(ИДЕНТИФИКАТОР_РЕПО)\n",
         "\n",
+        "print(\"Файлы репозитория:\")\n",
+        "for файл in СПИСОК_ФАЙЛОВ:\n",
+        "    print(файл)\n",
         "\n",
+        "ТОКЕНИЗАТОР = AutoTokenizer.from_pretrained(\n",
+        "    ИДЕНТИФИКАТОР_РЕПО,\n",
         "    trust_remote_code=True\n",
         ")\n",
         "\n",
+        "МОДЕЛЬ = AutoModelForCausalLM.from_pretrained(\n",
+        "    ИДЕНТИФИКАТОР_РЕПО,\n",
         "    torch_dtype=torch.float16,\n",
         "    device_map=\"auto\",\n",
         "    trust_remote_code=True\n",
         ")\n",
         "\n",
+        "МОДЕЛЬ.eval()\n"
       ]
     },
     {
       "metadata": {},
       "outputs": [],
       "source": [
+        "ЗАПРОС = \"Привет, как у тебя дела?\"\n",
         "\n",
+        "СООБЩЕНИЯ = [\n",
         "    {\n",
         "        \"role\": \"user\",\n",
+        "        \"content\": ЗАПРОС\n",
         "    }\n",
         "]\n",
         "\n",
+        "ТЕКСТ = ТОКЕНИЗАТОР.apply_chat_template(\n",
+        "    СООБЩЕНИЯ,\n",
         "    tokenize=False,\n",
         "    add_generation_prompt=True\n",
         ")\n",
         "\n",
+        "ВХОДЫ = ТОКЕНИЗАТОР(\n",
+        "    ТЕКСТ,\n",
         "    return_tensors=\"pt\"\n",
+        ").to(МОДЕЛЬ.device)\n",
         "\n",
         "with torch.no_grad():\n",
+        "    ОТВЕТЫ = МОДЕЛЬ.generate(\n",
+        "        **ВХОДЫ,\n",
         "        max_new_tokens=128,\n",
         "        temperature=0.7,\n",
         "        top_p=0.8,\n",
         "        top_k=20,\n",
         "        do_sample=True,\n",
+        "        eos_token_id=ТОКЕНИЗАТОР.eos_token_id\n",
         "    )\n",
         "\n",
+        "ОТВЕТ = ТОКЕНИЗАТОР.decode(\n",
+        "    ОТВЕТЫ[0][ВХОДЫ.input_ids.shape[-1]:],\n",
         "    skip_special_tokens=True\n",
         ")\n",
         "\n",
+        "print(ОТВЕТ)\n"
       ]
     }
   ],