Tencent releases Hy4 Flash, a 224B multimodal model with 12B active parameters
Tencent's multimodal Hy4 Flash has 224B total parameters, activates 12B per token, and is priced at $0.070 and $0.281 per million input and output tokens.
Tencent has released Hy4 Flash, a multimodal model with 224B total parameters. Of those, 12B are activated for each token it processes.
Put simply, the model stores 224B parameters but runs only 12B of them on any given token. Pricing is $0.070 per million input tokens and $0.281 per million output tokens. That gives developers a cheaper way to try Hy4's 1M-token context window.
What changed
Tencent released Hy4 Flash, a multimodal model with 224B total parameters and 12B active per token, priced at $0.070 per million input tokens and $0.281 per million output tokens.
Type: New capability · Applies to: Hy4 Flash
Why it matters
A low-cost sparse model with a 1M-token context could reduce the cost of long-document workloads. Independent benchmark results are not given in the source.
What you can do
Run a cost and quality comparison on your long-context workloads against your current model.
Original source
We have no link to the original source for this story yet.
Where we saw it
Latent Space
AI wrote this summary from the sources listed, and nobody reviewed it individually before publication. Check the original source before you rely on it. How it works