6 results for “GGUF”
Tencent open-sourced an extreme quantized GGUF build of its 770B Hunyuan Hy4 model, compressing 1.5TB weights down to 214GB. By varying precision from 1.31-bit on non-critical layers to 2+ bits on sensitive layers, the model fits on accessible multi-GPU nodes.
Tencent open-sourced Hy-MT1.5-1.8B packs that shrink a 3.3 GB translation model to as little as about 440 MB for offline use on ordinary phones. The 1.25-bit and 2-bit releases keep 33-language coverage while targeting on-device privacy and real-time translation without a network.
Alibaba's Qwen team released Qwen3.8-Flash-Next, a 125B-parameter multimodal MoE model with 6B activated parameters and 51B N-gram embeddings. The open-weight preview of Qwen4's architecture cuts training cost to one-ninth of its predecessor while leading coding and office-task benchmarks.
Tencent's Hunyuan team released Hy-MT2 on 21 May 2026 — a new translation model family with instruction-following templates, three sizes up to a 30B-A3B MoE variant, and open weights plus IFMTBench. This is the next generation after the April Hy-MT1.5 phone quantizations.
SenseTime fully open-sourced SenseNova U1, a native unified multimodal model that fuses vision and language without a vision encoder or VAE. The 8B-parameter release matches commercial models on infographic generation and adds native 4K output in its U1.5 update.
Liquid AI has released the LFM2.5 model family, a comprehensive suite of open-weight models designed to run entirely on edge devices without cloud. The LFM2.5 releases address three critical constraints that have limited on-device AI.