Instructions to use tokenaii/Horus-Taleeq-0.2B-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tokenaii/Horus-Taleeq-0.2B-Base with Transformers:

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

pipe = pipeline("text-generation", model="tokenaii/Horus-Taleeq-0.2B-Base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tokenaii/Horus-Taleeq-0.2B-Base")
model = AutoModelForCausalLM.from_pretrained("tokenaii/Horus-Taleeq-0.2B-Base", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tokenaii/Horus-Taleeq-0.2B-Base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tokenaii/Horus-Taleeq-0.2B-Base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tokenaii/Horus-Taleeq-0.2B-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/tokenaii/Horus-Taleeq-0.2B-Base

SGLang

How to use tokenaii/Horus-Taleeq-0.2B-Base 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 "tokenaii/Horus-Taleeq-0.2B-Base" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tokenaii/Horus-Taleeq-0.2B-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "tokenaii/Horus-Taleeq-0.2B-Base" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tokenaii/Horus-Taleeq-0.2B-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use tokenaii/Horus-Taleeq-0.2B-Base with Docker Model Runner:
```
docker model run hf.co/tokenaii/Horus-Taleeq-0.2B-Base
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Horus Taleeq 0.2 Base

Horus Taleeq 0.2 Base is an Arabic-only Small Language Model (SLM) with a decoder-only causal Transformer architecture, developed by TokenAI. TokenAI is a startup and nonprofit organization focused on open Arabic AI research and model development. It is a ready-to-build-on base checkpoint, not a finished instruction-tuned or production chat release.

Ownership and development

Company: TokenAI
Owner and founder: Assem Sabry
Lead developer: Assem Sabry
Hugging Face organization: tokenaii
Model repository: tokenaii/Horus-Taleeq-0.2-base
GitHub repository: tokenaii/Horus-Taleeq-0.2B

Model summary

Item	Value
Model family	Horus Taleeq
Model type	Arabic-only Small Language Model (SLM), decoder-only causal Transformer
Approximate parameters	0.2B (about 204.6M)
Primary language	Arabic
Intended text	Modern Standard Arabic and Arabic dialectal text
Egyptian Arabic	Targeted in the training program; not independently certified by this base checkpoint
Vocabulary	128,000-token Arabic-first SentencePiece tokenizer
Hidden size	640
Transformer layers	24
Attention heads	10
Key/value heads	2 (GQA)
MLP intermediate size	2,048
Activation	SiLU/SwiGLU-style Llama MLP
Normalization	RMSNorm, epsilon 1e-5
Position encoding	RoPE, theta 1,000,000
Embeddings	Tied input and output embeddings
Tensor type	bfloat16
Context configuration	4,096 tokens; initial pretraining sequences used 2,048 tokens

The uploaded checkpoint is the completed clean-repair-v1 base checkpoint. The later canonical-clean continuation is still a separate development run and is not included in this upload.

Training status

This repository is intended for the verified base checkpoint only. The original training ledger records approximately 6.0B tokenizer IDs across earlier training streams, but those historical streams are not yet a fully reproducible public corpus. A separate canonical-clean continuation is being evaluated and must not be represented as complete until its checkpoint and data manifest are published.

The current release therefore makes no claim of finished chat quality, factual benchmark leadership, or production readiness. Earlier conversation adapters are not part of this base repository.

The reproducibility baseline uses fused AdamW with betas (0.9, 0.95), epsilon 1e-8, weight decay 0.1, gradient clipping 1.0, CUDA bfloat16 autocast, and activation checkpointing. STAM is not the optimizer used for this verified base checkpoint.

Intended use

Arabic language-model research and evaluation
Continued pretraining and supervised fine-tuning experiments
Arabic and dialectal text generation research
Building a separate instruction/chat checkpoint after evaluation

This base model is ready for continued pretraining, instruction tuning, identity tuning, and downstream Arabic applications. It is not a safety-tuned assistant and should not be used as a production chatbot without additional alignment, safety, and quality evaluation.

Limitations

Base-model completion quality may be inconsistent and may repeat text.
It does not guarantee factual accuracy or reliable arithmetic.
Dialect fluency and dialect identification require held-out evaluation.
The model does not expose hidden chain-of-thought and should not be prompted to reveal one.
The historical training ledger and the later cleaned Parquet export are not identical; users must not claim full corpus reproducibility from this release.

Files in this repository

The planned base release contains the model configuration, bfloat16 weights, generation defaults, the matching 128K base tokenizer, the MIT license, and this model card. The role-special tokenizer created for future chat SFT is deliberately not presented as the tokenizer used by this base pretraining release. No SFT adapter, teacher outputs, private credentials, or raw training corpus is included.

Loading with Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "tokenaii/Horus-Taleeq-0.2-base"
tokenizer = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
    repo,
    torch_dtype="auto",
    device_map="auto",
)

prompt = "اكتب فقرة قصيرة عن أهمية القراءة."
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=120, do_sample=False)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Citation

@misc{tokenai_horus_taleeq_02_base_2026,
  title        = {Horus Taleeq 0.2 Base},
  author       = {Assem Sabry and TokenAI},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\\url{https://huggingface.co/tokenaii/Horus-Taleeq-0.2-base}},
  note         = {Arabic-first decoder-only base language model}
}

License

This model is released under the MIT License. Users remain responsible for complying with applicable laws, third-party data rights, and the terms of any data sources used in downstream training or evaluation.

Downloads last month: -

Safetensors

Model size

0.2B params

Tensor type

BF16

Model tree for tokenaii/Horus-Taleeq-0.2B-Base

Quantizations

1 model