<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Apple Machine Learning Research — FoundationIndex</title>
    <link>https://foundationindex.ustunozgur.com/s/apple-ml</link>
    <description>Apple&apos;s machine learning research and engineering division develops the frameworks, open research models, and foundation language and vision models that run across iOS, iPadOS, macOS, and Private Cloud Compute.</description>
    <language>en</language>
    <atom:link href="https://foundationindex.ustunozgur.com/s/apple-ml/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Apple Machine Learning Research · Release · 2026-06-08</title>
      <link>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-afm3-third-generation-2026</link>
      <guid>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-afm3-third-generation-2026</guid>
      <pubDate>Mon, 08 Jun 2026 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Apple introduced the third generation of Apple Foundation Models (AFM 3) alongside MLX optimizations for M5 GPU Neural Accelerators and Thunderbolt RDMA clusters, shipping AFM 3 Core and a sparse native multimodal AFM 3 Core Advanced on-device model (~20B total parameters activating 1B–4B per token across flash and active memory).&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://machinelearning.apple.com/research/introducing-third-generation-of-apple-foundation-models&quot;&gt;Introducing the Third Generation of Apple’s Foundation Models&lt;/a&gt;, &lt;a href=&quot;https://machinelearning.apple.com/research/mlx-distributed-inference&quot;&gt;Local and Distributed Large LLM Inference with MLX and RDMA on Mac Studio&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Apple Machine Learning Research · Release · 2025-06-09</title>
      <link>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-foundation-models-2025-fastvlm</link>
      <guid>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-foundation-models-2025-fastvlm</guid>
      <pubDate>Mon, 09 Jun 2025 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Apple detailed its 2025 foundation model upgrades and the Swift Foundation Models framework—combining 2-bit quantization-aware training and KV-cache sharing on-device with a Parallel-Track Mixture-of-Experts (PT-MoE) server backbone and the open-weight FastVLM (FastViTHD) hybrid vision encoder for low-latency high-resolution visual understanding.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://machinelearning.apple.com/research/apple-foundation-models-2025-updates&quot;&gt;Updates to Apple&amp;apos;s On-Device and Server Foundation Language Models&lt;/a&gt;, &lt;a href=&quot;https://machinelearning.apple.com/research/fastvlm-efficient-vision-encoding&quot;&gt;FastVLM: Efficient Vision Encoding for Vision Language Models&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Apple Machine Learning Research · Release · 2024-06-10</title>
      <link>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-foundation-models-2024</link>
      <guid>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-foundation-models-2024</guid>
      <pubDate>Mon, 10 Jun 2024 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Apple introduced the Apple Intelligence Foundation Language Models, pairing an approximately 3B-parameter on-device model (using grouped-query Attention, low-bit palletization, and dynamically swapped LoRA adapters) with a larger server foundation model running on Apple silicon Private Cloud Compute.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://machinelearning.apple.com/research/introducing-apple-foundation-models&quot;&gt;Introducing Apple’s On-Device and Server Foundation Models&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Apple Machine Learning Research · Release · 2024-04-22</title>
      <link>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-openelm-2024</link>
      <guid>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-openelm-2024</guid>
      <pubDate>Mon, 22 Apr 2024 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Following its MM1 multimodal pre-training study in March 2024, Apple released OpenELM (270M to 3B parameters) alongside complete CoreNet training logs, multiple Checkpoints, and MLX conversion utilities—using layer-wise scaling of attention heads and feed-forward dimensions to allocate parameters more efficiently across Transformer layers.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://machinelearning.apple.com/research/openelm&quot;&gt;OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework&lt;/a&gt;, &lt;a href=&quot;https://arxiv.org/abs/2404.14619&quot;&gt;OpenELM Technical Report (arXiv:2404.14619)&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Apple Machine Learning Research · Release · 2023-12-05</title>
      <link>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-mlx-framework-2023</link>
      <guid>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-mlx-framework-2023</guid>
      <pubDate>Tue, 05 Dec 2023 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Apple open-sourced MLX, a NumPy- and PyTorch-inspired array framework engineered for Apple silicon&amp;apos;s unified memory architecture so CPU and GPU operations share tensor buffers directly without host-to-device memory copies.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://github.com/ml-explore/mlx&quot;&gt;ml-explore/mlx: An array framework for Apple silicon&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Apple Machine Learning Research · Ecosystem · 2022-06-06</title>
      <link>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-neural-engine-transformers-2022</link>
      <guid>https://foundationindex.ustunozgur.com/s/apple-ml#apple-ml-neural-engine-transformers-2022</guid>
      <pubDate>Mon, 06 Jun 2022 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Apple Machine Learning Research published reference benchmarks and principles for deploying Transformer models on the Apple Neural Engine via Core ML, replacing nn.Linear projections with 4D spatial convolutions (bc1s) and splitting execution across CPU, GPU, and Neural Engine compute units to cut latency and peak memory on mobile devices.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://machinelearning.apple.com/research/neural-engine-transformers&quot;&gt;Deploying Transformers on the Apple Neural Engine&lt;/a&gt;&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
