{"version":1,"type":"story","url":"https://digestai.news/story/swift-1-5-qwen3-8-27b-updated-with-bf16-weights","json":"https://digestai.news/story/swift-1-5-qwen3-8-27b-updated-with-bf16-weights.json","markdown":"https://digestai.news/story/swift-1-5-qwen3-8-27b-updated-with-bf16-weights.md","slug":"swift-1-5-qwen3-8-27b-updated-with-bf16-weights","headline":"Swift 1.5 Qwen3.8-27b updated with BF16 weights","summary":"Swift 1.5 Qwen3.8-27b, a 28B image‑text‑to‑text model, was updated with full BF16 weights and various quantizations. Other options include 6B W4A16‑AWQ, 3B W4A16‑AutoRound, 28B Quark‑FP8‑dynamic‑AMD, 28B 5‑bit MLX, and a 27B GSQ‑RCO low‑bit GGUF with 1.12k downloads.\n\nThe update focuses on reducing thinking tokens and improving performance on agentic and coding tasks compared to Swift 1.0. Users can access the models via llama.cpp, LM Studio, Ollama, Apple Silicon Macs, NVIDIA Blackwell GPUs, and AMD GPUs, depending on the quantization and hardware support.","keyPoints":["Swift 1.5 Qwen3.8-27b is a 28B image‑text‑to‑text model updated with BF16 weights","Variants include 27B GGUF (3.3k downloads), 4‑bit MLX 28B (59 downloads), and 18B NVFP4 (102 downloads)","Available for llama.cpp, LM Studio, Ollama, Apple Silicon, NVIDIA Blackwell, and AMD GPUs"],"whyItMatters":"The release offers a range of quantized Swift 1.5 models that can run on diverse hardware, enabling developers to deploy efficient image‑text models in varied environments.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["ukisai"],"models":["Swift 1.5 Qwen3.8-27b"],"people":[]},"firstPublishedAt":"2026-09-26T07:28:27Z","updatedAt":"2026-09-26T07:28:27Z","sourceCount":1,"hasPrimarySource":true,"sources":[{"outlet":"huggingface.co","title":"Swift 1.5 27b: Swift Qwen just got faster","url":"https://huggingface.co/collections/ukisai/swift-15-27b","publishedAt":"2026-09-26T07:28:27Z","type":"primary","primary":true,"lead":true}],"sourceNotes":null,"discussions":[{"site":"Reddit","url":"https://www.reddit.com/r/LocalLLaMA/comments/1wqkb3v/swift_15_27b_swift_qwen_just_got_faster/","points":null}],"thread":null,"cite":{"text":"Digest AI, \"Swift 1.5 Qwen3.8-27b updated with BF16 weights\", 26 September 2026, https://digestai.news/story/swift-1-5-qwen3-8-27b-updated-with-bf16-weights","publisher":"Digest AI","title":"Swift 1.5 Qwen3.8-27b updated with BF16 weights","datePublished":"2026-09-26T07:28:27Z","url":"https://digestai.news/story/swift-1-5-qwen3-8-27b-updated-with-bf16-weights"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}