MikeTrendsTrends right now
AI: reposAPI

search

Small Language Models

Trends

  1. 1
    Microcontrollers now run a diffusion model and 289M-parameter LLM▼Microcontrollers now run a diffusion model and 289M LLM✉newsTechnologySoftware2 d ago

    Tiny microcontroller chips, traditionally limited to simple embedded tasks, can now run a diffusion model for image generation and a compact 289-million-parameter large language model. The news, highlighted by Adafruit and Open Source For You, points to rapid progress in on-device AI, letting small, low-power hardware perform generative tasks without cloud servers. Enthusiasts are discussing what this means for smart devices, robotics and offline AI applications.

  2. 2
    AWS Labs Launches Strands Decider 2B Open Source Decision Model▼AWS Strands Labs Releases Strands Decider 2B: An Open Source Decision Model That Picks Options in About 115 ms✉newsTechnologySoftware46 min ago

    AWS Strands Labs has released Strands Decider 2B, a new open source model designed to make quick decisions between options. According to the announcement, the model picks among choices in roughly 115 milliseconds, making it suited to latency-sensitive applications where larger language models would be too slow. The release adds to the growing set of small, specialized open models aimed at specific tasks rather than general-purpose reasoning.

  3. 3
    Quantized 27B Model Claimed to Match Frontier AI on Coding Task●A 27B Quantized LLM Is Said To Match Frontier AI Models In Just One Task From A Coding Benchmark, Making It A More Believable Claim✉newsTechnologyAI11 h ago

    A quantized 27-billion-parameter language model is reported to match frontier AI models on a single task from a coding benchmark. The narrow, specific nature of the claim makes it more believable than sweeping benchmark-superiority claims, but it also means the result says little about overall performance. Readers are debating how much weight such partial benchmark results deserve in judging open and smaller models.

  4. 4
    UC Santa Cruz's Adam Smith on local small language models●Adam Smith from UC Santa Cruz joins us to discuss local Small Language Models (SLMs) and building open, autonomous toolsMmastodonTechnologyAI210 h ago

    Adam Smith of UC Santa Cruz is discussing the case for running small language models locally rather than relying on large cloud providers. He presents BayLeaf AI, described as a counterplatform, along with the concept of "transagency" — a human-agent collaboration model he likens to the relationship between a driver and a car. The conversation also covers context distillation and practical approaches to building open, autonomous AI tools that users control themselves.

  5. 5
    TurboGPT trains tiny 22KiB transformer in 13 seconds●Show HN: TurboGPT: train 22KiB transformer in 13sYhnWarMiddle East412 d ago

    A developer known as lostmsu has released TurboGPT, an open-source project on GitHub that trains a compact 22KiB transformer model in roughly 13 seconds. The tool is drawing attention from machine learning enthusiasts interested in fast, lightweight training experiments that can run without large compute budgets.

Repos