AI 導讀
Mercury 3 是 Inception 推出的 diffusion LLM,單位秒可處理 1,068 tokens,速度為 GPT-6 Luna 與 Claude Haiku 5.5 的 4–8 倍。
推薦理由
Mercury 3 以 1,068 tokens/s 的速度,效能遠超 GPT-6 Luna 與 Claude Haiku 5.5,提供高效且高品質的推理體驗。
t
thehypedotnews twitter@thehypedotnews
·3 小時前
原文內容 (English)
inception dropped mercury 3 – a diffusion llm that runs at 1,068 tokens per second the specs: - inception says it is 4–8x faster than gpt-6 luna (~130 tok/s) and claude haiku 5.5 (~240 tok/s) - first in 8 of 12 benchmarks in inception's own table; behind on long context and
00:30 1/4 Today, we’re releasing Mercury 3. Mercury 3 is by far our most powerful model and the most intelligent diffusion LLM ever released. It outperforms GPT-6 Luna and Haiku 5.5 on quality while running at 1000+ tok/sec.
推文來源 · X (Twitter)在 X 參與討論
來源: thehypedotnews twitter (x.com)
在 X 查看推文