Swift 1.5 Qwen3.8-27b updated with BF16 weights
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.
Key points
- 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
The 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.
Swift 1.5 27b: Swift Qwen just got faster
huggingface.co · 26 September 2026
Loading the full article…
This text was published by huggingface.co. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
1source- Reddit discussionreddit.com
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- Liquid AI releases LFM2.5-VL-DSpark for faster vision-language model inference · 2 src
- Google releases Gemini 3.8 Flash TTS and Flash-Lite TTS with voice cloning and 100+ languages · 11 src
- Microsoft adds GPT-6 Sol, Luna and Astra to Foundry model lineup · 2 src
- Claude Opus 5.5 tops benchmark over OpenAI’s Astra and Fable 5.1 · 1 src
- Anthropic and OpenAI announce new AI models with lower prices on September 22, 2026 · 20 src
Comments
via GitHub Discussions