AI News HubLIVE
原文

RightNow-Arabic-0.5B-Turbo: An Open Sub-1B Arabic Language Model via Vocabulary Injection and Edge-First Deployment

This paper presents RightNow-Arabic-0.5B-Turbo, a 518M-parameter Arabic-specialized LLM built on Qwen2.5-0.5B using vocabulary injection and edge-first deployment. It achieves 35.9% mean accuracy on Arabic benchmarks, outperforming all same-class open models, and ties Falcon-H1-1.5B on COPA-ar at one-third the size. The quantized model is 398 MB and delivers 635 tokens/s on a single H100, enabling efficient edge deployment.

Article intelligence

InvestorsAdvanced

Key points

  • 518M-parameter Arabic LLM built on Qwen2.5-0.5B with vocabulary injection of 27,032 Arabic tokens.
  • Achieves 35.9% mean accuracy on three Arabic benchmarks, surpassing all same-class open-source models.
  • Matches Falcon-H1-1.5B on COPA-ar (58.4%) with one-third the parameters; recovers 67% of SILMA-9B's performance with 1/18 the parameters.
  • Quantized to 398 MB (q4_k_m), achieving 635 tokens/s at batch size 1 on a single H100 for edge deployment.

Why it matters

This matters because 518M-parameter Arabic LLM built on Qwen2.5-0.5B with vocabulary injection of 27,032 Arabic tokens.

Technical impact

May affect model selection, inference cost, product capability, and evaluation benchmarks.

[2605.28827] RightNow-Arabic-0.5B-Turbo: An Open Sub-1B Arabic Language Model via Vocabulary Injection and Edge-First Deployment

[Submitted on 10 Apr 2026]

Title:RightNow-Arabic-0.5B-Turbo: An Open Sub-1B Arabic Language Model via Vocabulary Injection and Edge-First Deployment

View a PDF of the paper titled RightNow-Arabic-0.5B-Turbo: An Open Sub-1B Arabic Language Model via Vocabulary Injection and Edge-First Deployment, by Jaber Jaber and 1 other authors

View PDF HTML (experimental)

Abstract:Open Arabic large language models split into two classes: sub-1B multilingual models that treat Arabic as an afterthought (Qwen2.5-0.5B, Falcon-H1-0.5B), and 7B-70B Arabic-specialized models that require a server to run (Jais, AceGPT, ALLaM, SILMA). The one published attempt at a sub-2B Arabic-specialized model, Kuwain-1.5B, never released its weights. We present RightNow-Arabic-0.5B-Turbo, a 518M-parameter Arabic-specialized decoder LLM built on Qwen2.5-0.5B. The pipeline adds 27,032 Arabic tokens via mean-subtoken initialization, continues pretraining on 504M Arabic tokens on 8xH100 with FSDP, FlashAttention varlen packing, and Liger fused kernels, then applies supervised fine-tuning on 129,116 Arabic instruction pairs with response-only loss masking, direct preference optimization on 6,750 Arabic preference pairs, and weight soup merging across three checkpoints. On three lm-evaluation-harness Arabic benchmarks (COPA-ar, Arabic HellaSwag, ArabicMMLU) the merged model reaches 35.9% mean accuracy, beats every same-class open model, ties Falcon-H1-1.5B on COPA-ar (58.4%) at one-third the size, and recovers 67% of SILMA-9B's mean at 1/18 the parameters. The edge build quantizes to 398 MB (q4_k_m) and delivers 635 tokens/s at batch size 1 on a single H100 via this http URL. All code (5,555 lines across 25 scripts), weights (bf16, int8, and four GGUF quantizations), and benchmark scripts are released at this https URL.

Comments: 12 pages, 7 tables, 4 figures, 1 algorithm. Weights: this https URL

Subjects:

Computation and Language (cs.CL); Machine Learning (cs.LG)

ACM classes: I.2.7

Cite as: arXiv:2605.28827 [cs.CL]

(or arXiv:2605.28827v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2605.28827

arXiv-issued DOI via DataCite

Submission history

From: Jaber Jaber [view email] [v1] Fri, 10 Apr 2026 00:56:50 UTC (66 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled RightNow-Arabic-0.5B-Turbo: An Open Sub-1B Arabic Language Model via Vocabulary Injection and Edge-First Deployment, by Jaber Jaber and 1 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-05

Change to browse by:

cs cs.LG

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)