Instructions to use leafspark/Mistral-Large-218B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use leafspark/Mistral-Large-218B-Instruct with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="leafspark/Mistral-Large-218B-Instruct")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("leafspark/Mistral-Large-218B-Instruct") model = AutoModelForCausalLM.from_pretrained("leafspark/Mistral-Large-218B-Instruct", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use leafspark/Mistral-Large-218B-Instruct with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "leafspark/Mistral-Large-218B-Instruct" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "leafspark/Mistral-Large-218B-Instruct", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/leafspark/Mistral-Large-218B-Instruct
- SGLang
How to use leafspark/Mistral-Large-218B-Instruct 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 "leafspark/Mistral-Large-218B-Instruct" \ --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": "leafspark/Mistral-Large-218B-Instruct", "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 "leafspark/Mistral-Large-218B-Instruct" \ --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": "leafspark/Mistral-Large-218B-Instruct", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use leafspark/Mistral-Large-218B-Instruct with Docker Model Runner:
docker model run hf.co/leafspark/Mistral-Large-218B-Instruct
model: update merge script config
Browse files
merge.py
CHANGED
|
@@ -22,8 +22,8 @@ def create_merge_plan(tensor_locations, layer_config):
|
|
| 22 |
# Special handling for specific weights
|
| 23 |
special_weights = {
|
| 24 |
"model.embed_tokens.weight": 1,
|
| 25 |
-
"lm_head.weight":
|
| 26 |
-
"model.norm.weight":
|
| 27 |
}
|
| 28 |
|
| 29 |
for slice_config in layer_config:
|
|
@@ -65,13 +65,18 @@ def create_merge_plan(tensor_locations, layer_config):
|
|
| 65 |
|
| 66 |
return merge_plan
|
| 67 |
|
| 68 |
-
def merge_layers(input_dir, output_dir, merge_plan):
|
| 69 |
output_tensors = {}
|
| 70 |
-
current_new_file_index = 1
|
| 71 |
max_file_index = max(item['new_file_index'] for item in merge_plan)
|
| 72 |
|
| 73 |
with tqdm(total=len(merge_plan), desc="Merging layers") as pbar:
|
| 74 |
-
for file_index in range(
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
for item in merge_plan:
|
| 76 |
if item['new_file_index'] == file_index:
|
| 77 |
input_file = os.path.join(input_dir, f"model-{item['original_file_index']:05d}-of-00051.safetensors")
|
|
@@ -81,7 +86,6 @@ def merge_layers(input_dir, output_dir, merge_plan):
|
|
| 81 |
pbar.update(1)
|
| 82 |
|
| 83 |
if output_tensors:
|
| 84 |
-
output_file = os.path.join(output_dir, f"model-{file_index:05d}-of-{max_file_index:05d}.safetensors")
|
| 85 |
save_file(output_tensors, output_file)
|
| 86 |
output_tensors = {}
|
| 87 |
|
|
@@ -92,6 +96,7 @@ def main():
|
|
| 92 |
parser.add_argument("input_dir", help="Directory containing input safetensors files")
|
| 93 |
parser.add_argument("output_dir", help="Directory for output safetensors files")
|
| 94 |
parser.add_argument("--dry-run", action="store_true", help="Perform a dry run and output merge plan")
|
|
|
|
| 95 |
args = parser.parse_args()
|
| 96 |
|
| 97 |
layer_config = [
|
|
@@ -116,8 +121,8 @@ def main():
|
|
| 116 |
print("Merge plan saved to merge_plan.json")
|
| 117 |
else:
|
| 118 |
os.makedirs(args.output_dir, exist_ok=True)
|
| 119 |
-
merge_layers(args.input_dir, args.output_dir, merge_plan)
|
| 120 |
print(f"Merged model saved to {args.output_dir}")
|
| 121 |
|
| 122 |
if __name__ == "__main__":
|
| 123 |
-
main()
|
|
|
|
| 22 |
# Special handling for specific weights
|
| 23 |
special_weights = {
|
| 24 |
"model.embed_tokens.weight": 1,
|
| 25 |
+
"lm_head.weight": 156,
|
| 26 |
+
"model.norm.weight": 156
|
| 27 |
}
|
| 28 |
|
| 29 |
for slice_config in layer_config:
|
|
|
|
| 65 |
|
| 66 |
return merge_plan
|
| 67 |
|
| 68 |
+
def merge_layers(input_dir, output_dir, merge_plan, start_file_index=1):
|
| 69 |
output_tensors = {}
|
|
|
|
| 70 |
max_file_index = max(item['new_file_index'] for item in merge_plan)
|
| 71 |
|
| 72 |
with tqdm(total=len(merge_plan), desc="Merging layers") as pbar:
|
| 73 |
+
for file_index in range(start_file_index, max_file_index + 1):
|
| 74 |
+
output_file = os.path.join(output_dir, f"model-{file_index:05d}-of-{max_file_index:05d}.safetensors")
|
| 75 |
+
|
| 76 |
+
if os.path.exists(output_file):
|
| 77 |
+
pbar.update(sum(1 for item in merge_plan if item['new_file_index'] == file_index))
|
| 78 |
+
continue
|
| 79 |
+
|
| 80 |
for item in merge_plan:
|
| 81 |
if item['new_file_index'] == file_index:
|
| 82 |
input_file = os.path.join(input_dir, f"model-{item['original_file_index']:05d}-of-00051.safetensors")
|
|
|
|
| 86 |
pbar.update(1)
|
| 87 |
|
| 88 |
if output_tensors:
|
|
|
|
| 89 |
save_file(output_tensors, output_file)
|
| 90 |
output_tensors = {}
|
| 91 |
|
|
|
|
| 96 |
parser.add_argument("input_dir", help="Directory containing input safetensors files")
|
| 97 |
parser.add_argument("output_dir", help="Directory for output safetensors files")
|
| 98 |
parser.add_argument("--dry-run", action="store_true", help="Perform a dry run and output merge plan")
|
| 99 |
+
parser.add_argument("--continue-from", type=int, default=1, help="Continue merging from this file index")
|
| 100 |
args = parser.parse_args()
|
| 101 |
|
| 102 |
layer_config = [
|
|
|
|
| 121 |
print("Merge plan saved to merge_plan.json")
|
| 122 |
else:
|
| 123 |
os.makedirs(args.output_dir, exist_ok=True)
|
| 124 |
+
merge_layers(args.input_dir, args.output_dir, merge_plan, start_file_index=args.continue_from)
|
| 125 |
print(f"Merged model saved to {args.output_dir}")
|
| 126 |
|
| 127 |
if __name__ == "__main__":
|
| 128 |
+
main()
|