Instructions to use kakaocorp/kanana-2-1.3b-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kakaocorp/kanana-2-1.3b-instruct with Transformers:

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

pipe = pipeline("text-generation", model="kakaocorp/kanana-2-1.3b-instruct", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("kakaocorp/kanana-2-1.3b-instruct", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use kakaocorp/kanana-2-1.3b-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kakaocorp/kanana-2-1.3b-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kakaocorp/kanana-2-1.3b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kakaocorp/kanana-2-1.3b-instruct

SGLang

How to use kakaocorp/kanana-2-1.3b-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 "kakaocorp/kanana-2-1.3b-instruct" \
    --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": "kakaocorp/kanana-2-1.3b-instruct",
		"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 "kakaocorp/kanana-2-1.3b-instruct" \
        --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": "kakaocorp/kanana-2-1.3b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kakaocorp/kanana-2-1.3b-instruct with Docker Model Runner:
```
docker model run hf.co/kakaocorp/kanana-2-1.3b-instruct
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

🤗 HF Models | 📕 Blog

News 🔥

2026/07/27: 🤗 Released kanana-2-3b, kanana-2-1.3b HF model weights.
2026/07/27: 📕 Published a blog post about the development of the Kanana-2 SLM series.

Introduction

We present Kanana-2 SLM, Kakao's second series of Small Language Models (SLMs), designed to deliver strong language capabilities while remaining compact and efficient for practical deployment. The series includes 3B, 1.3B, and 0.9B models. This release publicly includes the 3B model and the compressed 1.3B model, providing a balance between capability and efficiency for a wide range of applications. Kanana-2-3B was pretrained from scratch on TPU clusters and further improved through post-training with supervised fine-tuning and reinforcement learning, resulting in strong instruction-following and reasoning capabilities. Kanana-2-1.3B models are derived from Kanana-2-3B through a cascade pruning and distillation pipeline. To further improve deployment efficiency, they adopt Sliding Window Attention (SWA), enabling memory-efficient long-context inference with support for context lengths of up to 32K tokens while substantially reducing KV-cache memory requirements.

The Kanana-2 SLM release consists of the following four publicly available models:

Kanana-2-3B-Base — 3B pretrained base
Kanana-2-3B-Instruct — instruction-tuned 3B model
Kanana-2-1.3B-Base — compressed 1.3B pretrained base
Kanana-2-1.3B-Instruct — instruction-tuned 1.3B model for on-device deployment

No Kakao user data was used for either pre-training or post-training.

Highlights

Cascade Pruning & Distillation: Kanana-2-1.3B is built by progressively compressing Kanana-2-3B-Base (3B → 2B → 1.3B → 0.9B) through a cascade pruning and distillation pipeline.
Sliding Window Attention (SWA): Uses a 3:1 hybrid layout of sliding-window and full-attention layers. A sliding-window size of 1024 reduces per-token KV-cache reads, cutting KV-cache usage by up to ~72.7% at a 32K context length compared to a full-attention-only model. YaRN is applied to full-attention layers, while SWA layers retain RoPE, preserving long-range context without sacrificing local-attention efficiency.
Kanana-2 tokenizer: Improves Korean tokenization efficiency by over 30% compared to the previous generation.
Long context: Natively supports context lengths of up to 32,768 tokens.

Model Downloads

Model	Download
kanana-2-3b-base	🤗 HuggingFace
kanana-2-3b-instruct	🤗 HuggingFace
kanana-2-1.3b-base	🤗 HuggingFace
kanana-2-1.3b-instruct	🤗 HuggingFace

Performance

Base model evaluation results

Benchmark	Metric	Shot	kanana-2-3b-base	kanana-2-1.3b-base	Qwen3-1.7B-Base
General Tasks
MMLU	acc	5	62.77	56.49	62.32
MMLU-Pro	acc	5	36.54	29.87	37.08
BBH	acc	3	54.23	45.58	53.60
SimpleQA^†	acc	5	27.39	23.76	16.83
Mathematics Tasks
MATH	em	4	35.08	30.82	41.74
GSM8K	em	8	61.03	51.93	75.74
Coding Tasks
HumanEval	pass@1	0	55.96	51.33	45.31
MBPP	pass@1	3	50.95	44.86	53.85
Korean Tasks
KMMLU	acc	5	47.92	44.17	43.08
KoSimpleQA^†	acc	5	32.50	28.50	13.40
HAE-RAE Bench (v1.0)	acc	5	80.66	75.34	55.54
MATH-Ko^‡	em	4	31.54	24.75	31.92
MBPP-Ko^§	pass@1	3	44.55	39.38	47.97
Long Context Tasks
RULER-32K	acc	0	67.50	55.72	69.01

† Evaluated in Multiple Choice Question Answering (MCQA) format with 10 options.

‡ Subsets from HRM8K (MATH, GSM8K).

§ Internally translated to Korean.

Instruct model evaluation results

Instruction-following, chat, tool-calling, code, math, and knowledge benchmarks for the Kanana-2 SLM series. Scores use greedy decoding (temperature 0.0, top-p 1.0, max 4096 tokens); the metric for each benchmark is listed in the Metric column.

Benchmark	Metric	kanana-2-3b-instruct	kanana-2-1.3b-instruct	Qwen3.5-2B	Qwen3-1.7B
Chat
MT-Bench^†	judge	7.15	6.83	6.87	6.98
KoMT-Bench^†	judge	6.92	6.54	5.21	5.29
Instruction Following
IFBench	prompt strict	33.33	34.69	24.83	18.33
IFEval	prompt strict	80.96	77.63	66.91	68.39
IHEval	pass@1	35.96	27.06	38.52	42.09
Tool Calling
BFCL-v3 (Live)^‡	pass@1	71.94	69.64	66.96	65.48
BFCL-v3 (Multi-Turn)^‡	pass@1	17.12	5.50	6.27	4.38
Code Generation
MBPP	pass@1	70.63	69.05	55.56	62.17
MBPP+	pass@1	60.05	60.85	46.83	52.65
Mathematics
GSM-Plus	pass@1	61.17	58.27	61.29	63.10
MATH-500	pass@1	61.20	61.40	67.80	72.00
Minerva Math	pass@1	24.44	22.43	35.18	27.21
Reasoning & Knowledge
MMLU-CoT	acc	61.09	60.37	69.01	66.02
KMMLU-CoT	acc	43.32	42.79	41.75	37.84
HAERAE-Bench (v1.0)-CoT	acc	43.75	44.89	27.84	27.84
KoSimpleQA	acc	22.29	17.81	3.21	2.82

† Evaluated using gpt-4o-2024-08-06 as the judge model.

‡ Live denotes the average score of 6 live benchmarks, and Multi-Turn the average score of 4 multi-turn benchmarks.

Deployment

kanana-2-1.3b-instruct uses a custom hybrid attention architecture (Kanana2TinyForCausalLM — a Qwen3 backbone with a 3:1 SWA/full-attention layout and per-layer-type RoPE), shipped as remote code in the repository.

Because the modeling code is loaded from the repository, serving requires transformers >= 4.57 and the --trust-remote-code flag. For SGLang, use --attention-backend triton so the hybrid sliding-window attention is handled correctly.

vLLM

vLLM is a fast and memory-optimized engine designed for high-performance LLM inference and serving.

vllm serve kakaocorp/kanana-2-1.3b-instruct \
    --tensor-parallel-size 1 \
    --max-model-len 32768 \
    --trust-remote-code \
    --enable-auto-tool-choice \
    --tool-call-parser qwen3_coder

SGLang

SGLang is a high-efficiency framework for serving LLMs and VLMs, enabling easy deployment of OpenAI-compatible API servers.

For SGLang, the model is served through the stock Qwen3ForCausalLM path instead of the remote-code Kanana2TinyForCausalLM class. This requires two files shipped in the sglang/ directory of this repository:

sglang/config.json — a Qwen3-flavored config (architectures: ["Qwen3ForCausalLM"], model_type: qwen3, no auto_map) that keeps the hybrid-attention fields (layer_types, sliding_window, per-layer-type rope_parameters). Use it in place of the default config.json when serving with SGLang.
sglang/qwen3.py — a patched model definition that overrides the installed sglang/srt/models/qwen3.py.

python3 -m sglang.launch_server \
    --model-path kakaocorp/kanana-2-1.3b-instruct \
    --tp 1 \
    --context-length 32768 \
    --attention-backend triton \
    --trust-remote-code \
    --tool-call-parser qwen3_coder

Recommended: sglang==0.5.1.

With the Qwen3-flavored sglang/config.json the model itself needs no remote code, but --trust-remote-code is passed so the tokenizer/config are loaded without prompting.

Use triton or fa3 for the attention backend. Avoid flashinfer — it appears to have an issue with this model and causes significant accuracy/throughput degradation.

Kanana-2-1.3B shares the Qwen3 backbone but adds a 3:1 SWA/full hybrid attention layout and per-layer-type RoPE, which SGLang's stock Qwen3 model does not handle. The patched sglang/qwen3.py changes the decoder layer to:

Per-layer-type RoPE — for each layer, read the entry in config.rope_parameters matching config.layer_types[layer_id] and build that layer's own rotary embedding from it: full_attention uses YaRN (factor=40, original_max_position_embeddings=4096) and sliding_attention uses default RoPE (rope_theta=10000). A layer type with no matching entry falls back to no RoPE.
Hybrid sliding-window attention — layers typed sliding_attention run RadixAttention with sliding_window_size = config.sliding_window - 1 (SGLang uses an exclusive window, HF an inclusive one), while full_attention layers use full causal attention.

License

The model weights are released under the KananaOpenLicense.

Citation

@misc{kanana2slm2026,
  title = {Kanana-2 SLM},
  author = {Kanana LLM},
  year = {2026},
  url = {https://huggingface.co/collections/kakaocorp/kanana-2-slm}
}