Text Generation
Transformers
Safetensors
English
gemma4
image-text-to-text
function-calling
tool-use
bfcl
cloudsurf
qlora
gemma-4
conversational
Instructions to use cloudsurf-software/CloudSurf-4B-FC with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use cloudsurf-software/CloudSurf-4B-FC with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="cloudsurf-software/CloudSurf-4B-FC") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("cloudsurf-software/CloudSurf-4B-FC") model = AutoModelForMultimodalLM.from_pretrained("cloudsurf-software/CloudSurf-4B-FC", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use cloudsurf-software/CloudSurf-4B-FC with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "cloudsurf-software/CloudSurf-4B-FC" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cloudsurf-software/CloudSurf-4B-FC", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/cloudsurf-software/CloudSurf-4B-FC
- SGLang
How to use cloudsurf-software/CloudSurf-4B-FC 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 "cloudsurf-software/CloudSurf-4B-FC" \ --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": "cloudsurf-software/CloudSurf-4B-FC", "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 "cloudsurf-software/CloudSurf-4B-FC" \ --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": "cloudsurf-software/CloudSurf-4B-FC", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use cloudsurf-software/CloudSurf-4B-FC with Docker Model Runner:
docker model run hf.co/cloudsurf-software/CloudSurf-4B-FC
Add adapter merge utility referenced by the card
Browse files
scripts/fastloop_merge_adapter.py
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python3
|
| 2 |
+
"""fastloop_merge_adapter.py <base-dir> <adapter-dir> <merged-dir> — re-create the
|
| 3 |
+
BF16 merged checkpoint from a kept adapter (the ladder runners delete merged/
|
| 4 |
+
after eval; when an eval fails for plumbing reasons this rebuilds it in ~5 min
|
| 5 |
+
instead of a 2 h retune). Same merge law as fastloop_smoke_train.py M gate."""
|
| 6 |
+
import os, shutil, sys, torch
|
| 7 |
+
from peft import PeftModel
|
| 8 |
+
from transformers import AutoConfig, AutoModelForCausalLM, AutoModelForImageTextToText, AutoTokenizer
|
| 9 |
+
base, adapter, merged_dir = sys.argv[1:4]
|
| 10 |
+
cfg = AutoConfig.from_pretrained(base)
|
| 11 |
+
cls = AutoModelForImageTextToText if hasattr(cfg, "text_config") else AutoModelForCausalLM
|
| 12 |
+
m = cls.from_pretrained(base, device_map="auto", dtype=torch.bfloat16)
|
| 13 |
+
m = PeftModel.from_pretrained(m, adapter).merge_and_unload()
|
| 14 |
+
if os.path.isdir(merged_dir): shutil.rmtree(merged_dir)
|
| 15 |
+
m.save_pretrained(merged_dir, safe_serialization=True)
|
| 16 |
+
AutoTokenizer.from_pretrained(base).save_pretrained(merged_dir)
|
| 17 |
+
for f in os.listdir(base):
|
| 18 |
+
if f.endswith(".index.json"): continue
|
| 19 |
+
if f.endswith((".jinja", ".json", ".py", ".txt", ".model")) and not os.path.exists(os.path.join(merged_dir, f)):
|
| 20 |
+
shutil.copy(os.path.join(base, f), merged_dir)
|
| 21 |
+
st = [f for f in os.listdir(merged_dir) if f.endswith(".safetensors")]
|
| 22 |
+
print(f"MERGE PASS: {len(st)} shards -> {merged_dir}" if st else "MERGE FAIL")
|
| 23 |
+
sys.exit(0 if st else 1)
|