Instructions to use liu-nlp/Viking-7B-smol-smoltalk-sv with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use liu-nlp/Viking-7B-smol-smoltalk-sv with Transformers:

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

pipe = pipeline("text-generation", model="liu-nlp/Viking-7B-smol-smoltalk-sv")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("liu-nlp/Viking-7B-smol-smoltalk-sv")
model = AutoModelForCausalLM.from_pretrained("liu-nlp/Viking-7B-smol-smoltalk-sv", device_map="auto")
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 Settings

vLLM

How to use liu-nlp/Viking-7B-smol-smoltalk-sv with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "liu-nlp/Viking-7B-smol-smoltalk-sv"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "liu-nlp/Viking-7B-smol-smoltalk-sv",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/liu-nlp/Viking-7B-smol-smoltalk-sv

SGLang

How to use liu-nlp/Viking-7B-smol-smoltalk-sv 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 "liu-nlp/Viking-7B-smol-smoltalk-sv" \
    --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": "liu-nlp/Viking-7B-smol-smoltalk-sv",
		"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 "liu-nlp/Viking-7B-smol-smoltalk-sv" \
        --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": "liu-nlp/Viking-7B-smol-smoltalk-sv",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use liu-nlp/Viking-7B-smol-smoltalk-sv with Docker Model Runner:
```
docker model run hf.co/liu-nlp/Viking-7B-smol-smoltalk-sv
```

jekunz commited on Aug 13, 2025

Commit

26b86d9

verified ·

1 Parent(s): 1cf485e

Training in progress, step 50000

Browse files

Files changed (8) hide show

README.md +58 -0
config.json +1 -1
model-00001-of-00004.safetensors +1 -1
model-00002-of-00004.safetensors +1 -1
model-00003-of-00004.safetensors +1 -1
model-00004-of-00004.safetensors +1 -1
model.safetensors.index.json +1 -0
training_args.bin +2 -2

README.md ADDED Viewed

	@@ -0,0 +1,58 @@

+---
+base_model: LumiOpen/Viking-7B
+library_name: transformers
+model_name: Viking-7B-smol-smoltalk-sv
+tags:
+- generated_from_trainer
+- sft
+- trl
+licence: license
+---
+# Model Card for Viking-7B-smol-smoltalk-sv
+This model is a fine-tuned version of [LumiOpen/Viking-7B](https://huggingface.co/LumiOpen/Viking-7B).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+## Quick start
+```python
+from transformers import pipeline
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="liu-nlp/Viking-7B-smol-smoltalk-sv", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+## Training procedure
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/jenny-kunz-liu/huggingface/runs/zvs1omc5)
+This model was trained with SFT.
+### Framework versions
+- TRL: 0.21.0
+- Transformers: 4.55.1
+- Pytorch: 2.8.0
+- Datasets: 4.0.0
+- Tokenizers: 0.21.4
+## Citations
+Cite TRL as:
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

config.json CHANGED Viewed

@@ -24,7 +24,7 @@
   "rope_theta": 10000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.52.4",
   "untie_embeddings_and_output_weights": true,
   "use_cache": true,
   "vocab_size": 131072

   "rope_theta": 10000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.1",
   "untie_embeddings_and_output_weights": true,
   "use_cache": true,
   "vocab_size": 131072

model-00001-of-00004.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c3795263c77d3ff227ab335e6ca7b3e66b61c5f72085566c5b521a64a4283ec
 size 4941047672

 version https://git-lfs.github.com/spec/v1
+oid sha256:f6797634a480ceec712a996b54d2bd90f3fb2da53608356ab6e9811eb1d7da7a
 size 4941047672

model-00002-of-00004.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2ee392ebb1bc30b581241a7fd227928bda1205d4950ee496066f3c177004e248
 size 4947390880

 version https://git-lfs.github.com/spec/v1
+oid sha256:4e201d7b0ce91da145e3bdfde9ffd5f715ff84a1865172ad8d8bf0a05c1adf0f
 size 4947390880

model-00003-of-00004.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e63b0455ee0d632688da35f5d357e3b4e5273ad8185b097590287f2dbaa6376
 size 4137880232

 version https://git-lfs.github.com/spec/v1
+oid sha256:e52fe83c4ec89afb37bed5733294105569c6ffeea70007a65ae1605d48d66307
 size 4137880232

model-00004-of-00004.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:baa88ce8c9a40ee950fc995ecb7425b7bfaf6cf1bb310f1bad5100ef20353e9f
 size 1073741952

 version https://git-lfs.github.com/spec/v1
+oid sha256:951f7a781903b5c48e92025e6e29ce2ea5417e128bd45d04fbcdcf668de7f89c
 size 1073741952

model.safetensors.index.json CHANGED Viewed

@@ -1,5 +1,6 @@
 {
   "metadata": {
     "total_size": 15100026880
   },
   "weight_map": {

 {
   "metadata": {
+    "total_parameters": 7550013440,
     "total_size": 15100026880
   },
   "weight_map": {

training_args.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:51f51d4404aecb3d521c551a0276c6d9e2f3abb1502a8a55174efbc5ac811304
-size 5624

 version https://git-lfs.github.com/spec/v1
+oid sha256:51dad3a86f75c1ff687037b47a2fc8c012ac4d54c4995a14e56b8920cd21eb72
+size 6161