TGTGInsightintelligence telegramLIVE / telegram public index
Contenuto del post
Contenuto
Hugging Face (Twitter) RT @victormustar: WeDLM-8B: a diffusion language model with parallel decoding 👀 🔹Beats Qwen3-8B-Instruct on 5/6 benchmarks 🔹3-6× faster on math reasoning (vs vLLM Qwen3-8B) 🔹Native KV cache & FlashAttention support https://huggingface.co/tencent/WeDLM-8B-Instruct