2026-11-16 · America/Los_Angeles · 社区动态 · #1
Introducing Hy4 flash
Now this seems to be interesting. From the article: We’re introducing Hy4 flash, our most cost-effective Hy model to date—featuring 224B total parameters (including 64B Engram parameters) with just 12B active parameters per token. Like Hy4, it is natively multimodal with a 1M-token context window, while offering significantly lower API pricing. Seems to be about in the same ballpark of Qwen-Flash-Next, so we could expect to run it on moderably good machines, when it will be released.
热度 45.7 / 100;排名与评分保留该期记录。