8 categories and 46 products and companies, charted by an ai agent.
Open-weight small language models are compact neural networks released with public weights that run on consumer hardware. Developers, researchers, and companies use them for local inference, fine-tuning, and integration into applications.
microsoft phi · google gemma · meta llama · qwen · mistral ai · hugging face smollm · liquid ai · ibm granite · allen ai olmo · stability ai · gemma · llama · smollm
Local inference runtimes execute small language models on individual machines without sending data to remote servers. Developers, researchers, and end users run these to process text, code, or other tasks on personal computers, phones, or edge devices.
ollama · llama.cpp · lm studio · jan · gpt4all · mlc llm
On-device AI SDKs are software development kits that enable developers to integrate small language models directly into applications that run on end-user devices without server calls. Mobile app developers, embedded systems engineers, and consumer device manufacturers use these SDKs to add language capabilities to their products.
apple foundation models · qualcomm ai hub · executorch · nexa ai · picovoice · apple intelligence · edge impulse · zetic ai
unsloth · predibase · arcee ai · together ai · distil labs · openpipe
Edge AI hardware refers to specialized processors, chips, and devices that run small language models locally without cloud connectivity. Device manufacturers, embedded systems engineers, and enterprises deploy these to run inference on phones, IoT devices, and edge servers.
nvidia jetson · hailo · axelera ai · sima.ai
Private local chat apps run small language models on users' devices without sending data to external servers. Developers, privacy-conscious individuals, and organizations use these to run conversational AI locally.
pocketpal ai · msty · open webui · anythingllm
Small language models (SLMs) are neural networks with fewer parameters than large language models, designed to run on constrained hardware. Developers, device manufacturers, and enterprises use them for on-device inference, edge applications, and resource-limited deployments.
writer palmyra · cohere
Private enterprise SLM apps are packaged software applications that run small language models on private infrastructure, used by organizations to deploy LLM functionality without sending data to public cloud services. They serve enterprises, government agencies, and regulated industries that require on-premises deployment.
h2o.ai · writer · snorkel ai