Instructions to use FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch with Transformers:

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

pipe = pipeline("text-generation", model="FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch")
model = AutoModelForCausalLM.from_pretrained("FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch

SGLang

How to use FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch 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 "FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch" \
    --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": "FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch",
		"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 "FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch" \
        --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": "FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch with Docker Model Runner:
```
docker model run hf.co/FLOPS-Squared/KeystoneFuse-B-FuserWidth-32-Instruct-PyTorch
```

nosuchjihyun commited on Jun 6, 2025

Commit

4d664b2

verified ·

1 Parent(s): ace55df

Upload folder using huggingface_hub

Browse files

Files changed (3) hide show

model.safetensors +1 -1
tokenizer.json +6 -1
tokenizer_config.json +4 -0

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50b4b8c859f3d8fb4f5bdea5ccbb59b06d2dca1017564606af59a1e0953f1b1c
 size 503914648

 version https://git-lfs.github.com/spec/v1
+oid sha256:584204b6484b226852feca624d854e6429346db32a8765e263557a7de7bc35ed
 size 503914648

tokenizer.json CHANGED Viewed

@@ -1,6 +1,11 @@
 {
   "version": "1.0",
-  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

 {
   "version": "1.0",
+  "truncation": {
+    "direction": "Right",
+    "max_length": 2014,
+    "strategy": "LongestFirst",
+    "stride": 0
+  },
   "padding": null,
   "added_tokens": [
     {

tokenizer_config.json CHANGED Viewed

@@ -256,8 +256,12 @@
   "cls_token": "<classification>",
   "eos_token": "<eos>",
   "extra_special_tokens": {},
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",
   "tokenizer_class": "GPT2Tokenizer",
   "unk_token": "<unk>"
 }

   "cls_token": "<classification>",
   "eos_token": "<eos>",
   "extra_special_tokens": {},
+  "max_length": 2014,
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",
+  "stride": 0,
   "tokenizer_class": "GPT2Tokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
   "unk_token": "<unk>"
 }