Instructions to use lightonai/Qwen3-8B-SW with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lightonai/Qwen3-8B-SW with Transformers:

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

pipe = pipeline("text-generation", model="lightonai/Qwen3-8B-SW")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lightonai/Qwen3-8B-SW")
model = AutoModelForCausalLM.from_pretrained("lightonai/Qwen3-8B-SW")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use lightonai/Qwen3-8B-SW with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lightonai/Qwen3-8B-SW"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lightonai/Qwen3-8B-SW",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lightonai/Qwen3-8B-SW

SGLang

How to use lightonai/Qwen3-8B-SW 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 "lightonai/Qwen3-8B-SW" \
    --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": "lightonai/Qwen3-8B-SW",
		"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 "lightonai/Qwen3-8B-SW" \
        --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": "lightonai/Qwen3-8B-SW",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use lightonai/Qwen3-8B-SW with Docker Model Runner:
```
docker model run hf.co/lightonai/Qwen3-8B-SW
```

Qwen3-8B-SW

Qwen3-8B-SW is a native reasoning model fine-tuned from Qwen/Qwen3-8B-Base to reason in Swahili. This model produces its entire reasoning trace in Swahili before delivering the final answer in Swahili.

It is released alongside the paper Rethinking the Multilingual Reasoning Gap with Layer Swap.

Model details

Base model: Qwen/Qwen3-8B-Base
Language: Swahili (CoT and answer)
Training: Full SFT, ~10B tokens, 2 epochs
Context length: 32,768 tokens
Dataset: lightonai/Dolci-Think-SFT-32B-Multilingual (Swahili split).

The model was trained on data derived from allenai/Dolci-Think-SFT-32B, released under the ODC-BY-1.0 license.

Related models

This model is part of a Swahili specialist trio designed to study the native reasoning gap:

Model	CoT language	Description
`lightonai/Qwen3-8B-SW`	Swahili	Native reasoning specialist
`lightonai/Qwen3-8B-SW-Swap`	Swahili	Layer Swap: middle layers (L13–L22) of `Qwen3-8B-EN` transplanted into `Qwen3-8B-SW`
`lightonai/Qwen3-8B-SW-Pivot-EN`	English	Same Swahili Q&A pairs, but CoT in English
`lightonai/Qwen3-8B-EN`	English	English specialist

Evaluation

All scores are mean accuracy (%) on the Swahili version of each benchmark, with sample standard deviation across runs. AIME 24/25 is averaged over 30 runs; the others over 10 runs, using the recommended generation parameters.

Model	MGSM-Rev2	Global-MMLU-Lite	GPQA-Diamond	AIME 24/25	HumanEvalPlus	Average
`Qwen3-8B-SW`	93.16	61.98	49.39	47.67	82.69	66.98
`Qwen3-8B-SW-Swap`	96.12	64.10	49.29	50.33	85.62	69.09
`Qwen3-8B-SW-Pivot-EN`	89.68	66.00	52.73	59.67	84.50	70.52
`Qwen3-8B-EN`	35.88	33.88	36.82	24.78	58.44	37.96

Benchmarks used:

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "lightonai/Qwen3-8B-SW"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto")

messages = [{"role": "user", "content": "Suluhisha: 24 × 17 = ?"}]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True).to(model.device)
outputs = model.generate(inputs, max_new_tokens=32768, temperature=1.0, top_p=0.95, top_k=20)
print(tokenizer.decode(outputs[0][inputs.shape[-1]:], skip_special_tokens=True))

Recommended sampling: temperature=1.0, top_p=0.95, top_k=20, min_p=0.

Citation

If you find our work helpful, feel free to give us a cite.

@misc{lasbordes2026rethinking,
  title        = {Rethinking the Multilingual Reasoning Gap with Layer Swap},
  author       = {Lasbordes, Maxence and Chatelain, Amélie and Seddah, Djamé},
  year         = {2026},
  eprint       = {2605.26735},
  archivePrefix= {arXiv},
  primaryClass = {cs.CL}
}