Tencent's Hy-SimulMT-30B-A3B-FP8 translation model trends on Hugging Face
tencent · hf · 2026-10-28
Tencent's Hy-SimulMT-30B-A3B-FP8 is trending on Hugging Face. Built on the Hyv3 architecture, the 30B-A3B model targets translation/text generation across zh, en, fr, pt and es, shipped in FP8 quantization with transformers and safetensors support.
More from Models
- GPT-5.5 passes 46.3% of clinical tasks once, but only 28% across three runs — jonc101x · 2026-10-28
- Liquid AI's 1.7B model claimed to beat models 2-5x its size on phones and laptops — rickasaurus · 2026-10-28
- GitHub Copilot Verified Proficiency badges roll out, developer shows off Proficient tier — PaulShellDev · 2026-10-28
- New RWE-bench: Best AI passes just 37% of real engineering tasks — ZeroCool86 · 2026-10-28
- Artificial Analysis' new benchmark table called 'absolute cinema' for its naming and DeepSeek column — eliebakouch · 2026-10-28
- Researcher roasts a new model's benchmark table: "the more you look, the worse it gets" — eliebakouch · 2026-10-28