Small Language Model

TII has released Falcon-ASR, a 1.6B speech recognition model for Arabic with an Emirati focus. It reports 20.92% average WER on 6 Arabic test sets, but weights and an API are not yet available.

Nace.AI has open-sourced Drex 1.5, a 9B decision model that returns a probability for every option in 1 forward pass. It scores 58.08 on Decision Index 0.3.1 and reads up to 128K tokens.

Unsloth's October 6 security overview explains how Studio checks code, weights, packages and tools before anything runs. Custom model code is scanned and approval is bound to its fingerprint. Flagged weight files are blocked in the load path, package-content findings fail CI, and tools run in probed OS sandboxes. Here is what each checkpoint decides, and what it does not cover.

Alibaba's Qwen went from an invite-only chatbot in April 2023 to a 2.4-trillion-parameter open-weight model in August 2026. This is the full story, release by release: every major model, its key feature, and how its license changed. Each claim links to its source.

This tutorial provides a complete workflow for building a compact, reasoning-focused language model. By streaming the SupraLabs reasoning corpus from Hugging Face, we apply quality filters and curate data for Supervised Fine-Tuning (SFT). Using SmolLM2-135M-Instruct and LoRA, we demonstrate an end-to-end pipeline—from dataset analysis and heuristic cleaning to efficient training and inference—enabling the development of specialized small models without excessive resource requirements

NVIDIA researchers have introduced Star Elastic, a post-training method that embeds multiple nested reasoning models — at 30B, 23B, and 12B parameter scales — inside a single checkpoint, eliminating the need for separate training runs or stored model weights per variant. Built on the Nemotron Elastic framework and applied to Nemotron Nano v3, the method trains all three variants in a single 160B-token run, achieving a 360× token reduction compared to pretraining each model from scratch. Beyond training efficiency, Star Elastic introduces elastic budget control — a novel inference scheme that uses a smaller submodel for the thinking phase and the full model for the final answer — delivering up to 16% higher accuracy and 1.9× lower latency compared to standard budget control, while nested FP8 and NVFP4 checkpoints bring the full model family within reach of RTX-class GPUs.

Run Google’s latest omni-capable open models faster on NVIDIA RTX AI PCs, from NVIDIA Jetson Orin Nano, GeForce RTX desktops to the new DGX...

Mistral AI has released Mistral Small 4, a new model in the Mistral Small family designed to consolidate several previously separate capabilities into a...

Microsoft has released Phi-4-reasoning-vision-15B, a 15 billion parameter open-weight multimodal reasoning model designed for image and text tasks that require both perception and selective...

Alibaba’s Qwen team has released the Qwen3.5 Small Model Series, a collection of Large Language Models (LLMs) ranging from 0.8B to 9B parameters. While...

Brain-computer interfaces (BCIs) are finally having their 'foundation model' moment. Zyphra, a research lab focused on large-scale models, recently released ZUNA, a 380M-parameter foundation...

Liquid AI has introduced LFM2.5, a new generation of small foundation models built on the LFM2 architecture and focused at on device and edge...

How can a small model learn to solve tasks it currently fails at, without rote imitation or relying on a correct rollout? A team...

Small models are often blocked by poor instruction tuning, weak tool use formats, and missing governance. IBM AI team released Granite 4.0 Nano, a...

Can a compact late interaction retriever index once and deliver accurate cross lingual search with fast inference? Liquid AI released LFM2-ColBERT-350M, a compact late...

Anthropic released Claude Haiku 4.5, a latency-optimized “small” model that delivers similar levels of coding performance to Claude Sonnet 4 while running more than...

Do you actually need a giant VLM when dense Qwen3-VL 4B/8B (Instruct/Thinking) with FP8 runs in low VRAM yet retains 256K→1M context and the...

How much capability can a sparse 8.3B-parameter MoE with a ~1.5B active path deliver on your phone without blowing latency or memory? Liquid AI...

Researchers from Cornell and Google introduce a unified Regression Language Model (RLM) that predicts numeric outcomes directly from code strings—covering GPU kernel latency, program...

EmbeddingGemma is Google’s new open text embedding model optimized for on-device AI, designed to balance efficiency with state-of-the-art retrieval performance. How compact is EmbeddingGemma compared...

Table of contents1. Regulatory and Risk Posture2. Capability vs. Cost, Latency, and Footprint3. Security and Compliance Trade-offs4. Deployment Patterns5. Decision Matrix (Quick Reference)6. Concrete...

Smaller Models with Smarter Performance and 256K Context Support Alibaba’s Qwen team has introduced two powerful additions to its small language model lineup: Qwen3-4B-Instruct-2507 and...

OpenAI has just sent seismic waves through the AI world: for the first time since GPT-2 hit the scene in 2019, the company is...

Mistral AI, in collaboration with All Hands AI, has released updated versions of its developer-focused large language models under the Devstral 2507 label. The...

Hugging Face just released SmolLM3, the latest version of its “Smol” language models, designed to deliver strong multilingual reasoning over long contexts using a...

Language processing in enterprise environments faces critical challenges as business workflows increasingly depend on synthesising information from diverse sources, including internal documentation, code repositories,...

AI models today are expected to handle complex tasks such as solving mathematical problems, interpreting logical statements, and assisting with enterprise decision-making. Building such...

As language models continue to grow in size and complexity, so do the resource requirements needed to train and deploy them. While large-scale models...

Today, Meta AI announced the release of its latest generation multimodal models, Llama 4, featuring two variants: Llama 4 Scout and Llama 4 Maverick....

In the realm of artificial intelligence, multilingual speech recognition and translation have become essential tools for facilitating global communication. However, developing models that can...

In today’s dynamic AI landscape, developers and organizations face several practical challenges. High computational demands, latency issues, and limited access to truly adaptable open-source...

In today’s rapidly evolving digital landscape, the need for accessible, efficient language models is increasingly evident. Traditional large-scale models have advanced natural language understanding...

Large language models (LLMs) leverage deep learning techniques to understand and generate human-like text, making them invaluable for various applications such as text generation,...

For many years, organizations in the MENA region have encountered difficulties when integrating AI solutions that truly understand the Arabic language. Traditional models have...

In today’s rapidly evolving technological landscape, developers and organizations often grapple with a series of practical challenges. One of the most significant hurdles is...

The integration of visual and textual data in artificial intelligence presents a complex challenge. Traditional models often struggle to interpret structured visual documents such...

In the evolving landscape of artificial intelligence, integrating vision and language capabilities remains a complex challenge. Traditional models often struggle with tasks requiring a...

The advancements in large language models (LLMs) have significantly enhanced natural language processing (NLP), enabling capabilities like contextual understanding, code generation, and reasoning. However,...

Text-to-speech (TTS) technology has emerged as a critical tool for bridging the gap between human and machine interaction. The demand for lifelike, emotionally resonant,...

The growing reliance on AI models for edge and mobile devices has underscored significant challenges. Balancing computational efficiency, model size, and multilingual capabilities remains...

Large language models (LLMs) like GPT-4, PaLM, Bard, and Copilot have made a huge impact in natural language processing (NLP). They can generate text,...

Microsoft has released Phi-4, a compact and efficient small language model, on Hugging Face under the MIT license. This decision highlights a shift towards...

Medical artificial intelligence (AI) is full of promise but comes with its own set of challenges. Unlike straightforward mathematical problems, medical tasks often demand...

The rapid development of Large Language Models (LLMs) has transformed natural language processing (NLP). Proprietary models like GPT-4 and Claude 3 have set high...

Large Language Models (LLMs) play a vital role in many AI applications, ranging from text summarization to conversational AI. However, evaluating these models effectively...

Large Language Models (LLMs) are the backbone of numerous applications, such as conversational agents, automated content creation, and natural language understanding tasks. Their effectiveness...

The integration of artificial intelligence into everyday life faces notable hurdles, particularly in multimodal understanding—the ability to process and analyze inputs across text, audio,...

The advancements in large language models (LLMs) have created opportunities across industries, from automating content creation to improving scientific research. However, significant challenges remain....

While multimodal models (LMMs) have advanced significantly for text and image tasks, video-based models remain underdeveloped. Videos are inherently complex, combining spatial and temporal...

Audio language models (ALMs) play a crucial role in various applications, from real-time transcription and translation to voice-controlled systems and assistive technologies. However, many...

Large Language Models (LLMs) have achieved remarkable advancements in natural language processing (NLP), enabling applications in text generation, summarization, and question-answering. However, their reliance...

Large language models (LLMs) are increasingly essential for enterprises, powering applications such as intelligent document processing and conversational AI. However, their adoption is often...

Large language models have made impressive strides in understanding natural language, solving programming tasks, and tackling reasoning challenges. However, their high computational costs and...

The ongoing advancement in artificial intelligence highlights a persistent challenge: balancing model size, efficiency, and performance. Larger models often deliver superior capabilities but require...

LG AI Research has released bilingual models expertizing in English and Korean based on EXAONE 3.5 as open source following the success of its...

Large language models (LLMs) have profoundly influenced natural language processing (NLP), excelling in tasks like text generation and language understanding. However, the Arabic language—with...

Snowflake recently announced the launch of Arctic Embed L 2.0 and Arctic Embed M 2.0, two small and powerful embedding models tailored for multilingual...

Ruliad AI released Deepthought-8B-LLaMA-v0.01-alpha, focusing on reasoning transparency and control. This model, built on LLaMA-3.1 with 8 billion parameters, is designed to offer sophisticated...

Understanding protein sequences and their functions has always been a challenging aspect of protein research. Proteins, often described as the building blocks of life,...

In recent years, there has been a growing demand for machine learning models capable of handling visual and language tasks effectively, without relying on...

The rapid growth in AI model sizes has brought significant computational and environmental challenges. Deep learning models, particularly language models, have expanded considerably in...

In today's increasingly interconnected world, effective communication across languages is essential. However, many natural language processing (NLP) models still struggle with less common languages....

Edge AI has long faced the challenge of balancing efficiency and effectiveness. Deploying Vision Language Models (VLMs) on edge devices is difficult due to...

Recent advancements in large language models (LLMs) have demonstrated significant capabilities in a wide range of applications, from solving mathematical problems to answering medical...

In the world of software development, there is a constant need for more intelligent, capable, and specialized coding language models. While existing models have...

AI's rapid rise has been driven by powerful language models, transforming industries from customer service to content creation. However, many languages, particularly those from...

In the rapidly evolving world of artificial intelligence and machine learning, the demand for powerful, flexible, and open-access solutions has grown immensely. Developers, researchers,...

In recent years, the surge in large language models (LLMs) has significantly transformed how we approach natural language processing tasks. However, these advancements are...

LLMWare.ai, a pioneer in deploying and fine-tuning Small Language Models (SLMs) announced today the launching of Model Depot in Hugging Face, one of the...

In the rapidly evolving world of finance, the demand for models that provide robust insights has never been greater. Traditional financial analysis requires an...

Despite rapid advancements in language technology, significant gaps in representation persist for many languages. Most progress in natural language processing (NLP) has focused on...

The rapid growth of large language models (LLMs) has brought significant advancements across various sectors, but it has also presented considerable challenges. Models such...

Artificial intelligence is advancing rapidly, but enterprises face many obstacles when trying to leverage AI effectively. Organizations require models that are adaptable, secure, and...

In the rapidly evolving world of AI, challenges related to scalability, performance, and accessibility remain central to the efforts of research communities and open-source...

High-performance AI models that can run at the edge and on personal devices are needed to overcome the limitations of existing large-scale models. These...

Zyphra has officially released Zamba2-7B, a state-of-the-art small language model that promises unprecedented performance in the 7B parameter range. This model outperforms existing competitors,...

In the ever-evolving world of artificial intelligence (AI), large language models have proven instrumental in addressing a wide array of challenges, from automating complex...

Google has launched the "gemma-2-2b-jpn-it" model, a new addition to its Gemma family of language models. The model is designed to cater specifically to...

The AI research organization Zyphra has recently unveiled two groundbreaking language models, Zamba2-1.2B-Instruct and Zamba2-2.7B-Instruct. These models are part of the Zamba2 series and...

Liquid AI has released its first series of Liquid Foundation Models (LFMs), ushering in a new generation of generative AI models. These models are...

Climate and weather prediction has experienced rapid advancements through machine learning and deep learning models. Researchers have started to rely on artificial intelligence (AI)...

The advancement of large language models (LLMs) in natural language processing has significantly improved various domains. As more complex models are developed, evaluating their...

AMD has recently introduced its new language model, AMD-135M or AMD-Llama-135M, which is a significant addition to the landscape of AI models. Based on...

Voyage AI is proud to announce the release of its new generation of embedding models, Voyage-3 and Voyage-3-Lite. The Voyage-3 and Voyage-3-Lite models are...

Small language models (SLMs) have become a focal point in natural language processing (NLP) due to their potential to bring high-quality machine intelligence to...

Multimodal models represent a significant advancement in artificial intelligence by enabling systems to process and understand data from multiple sources, like text and images....

The demand for customizable, open models that can run efficiently on various hardware platforms has grown, and Meta is at the forefront of catering...

The University of Washington and the Allen Institute for AI (Ai2) have recently made a significant contribution to the AI research community by releasing...

The Qwen team from Alibaba has recently made waves in the AI/ML community by releasing their latest series of large language models (LLMs), Qwen2.5....

The field of spoken dialogue systems has evolved significantly over the years, moving beyond simple voice-based interfaces to complex models capable of sustaining real-time...

Mistral AI recently announced the release of Mistral-Small-Instruct-2409, a new open-source large language model (LLM) designed to address critical challenges in artificial intelligence research...

Large Language Models (LLMs) have revolutionized natural language processing in recent years. The pre-train and fine-tune paradigm, exemplified by models like ELMo and BERT,...

Nvidia has unveiled its latest small language model, Nemotron-Mini-4B-Instruct, which marks a new chapter in the company’s long-standing tradition of innovation in artificial intelligence....

The Internet Integrity Initiative Team has made a significant stride in data privacy by releasing Piiranha-v1, a model specifically designed to detect and protect...

The release of Reader-LM-0.5B and Reader-LM-1.5B by Jina AI marks a significant milestone in small language model (SLM) technology. These models are designed to...

OpenBMB recently released the MiniCPM3-4B, the third-generation model in the MiniCPM series. This model marks a great step forward in the capabilities of smaller-scale...

Machine learning models, especially those designed for code generation, heavily depend on high-quality data during pretraining. This field has seen rapid advancement, with large...

Artificial intelligence (AI) has witnessed rapid advancements over the past decade, with significant strides in NLP, machine learning, and deep learning. Among the latest...

A critical challenge in training large language models (LLMs) for reasoning tasks is identifying the most compute-efficient method for generating synthetic data that enhances...

Cartesia AI has made a notable contribution with the release of Rene, a 1.3 billion-parameter language model. This open-source model, built upon a hybrid...