DeepSeek V4.1 Flash could run on under 10 Cerebras wafers
An analysis argues only 500B of DeepSeek V4.1 Flash's weights need to sit on wafer with 192B of embeddings off-chip, so Cerebras could serve it efficiently on fewer than 10 wafers.
True, Cerebras needs 60 wafers for K3. That is because K3 is huge model with 2.8 trillion parameters. With 1.5TB in storage. But new architectures allow it to cut down that count to just 10 with total hardware cost approaching a single NVL72, imho. Most near Mythos capable models for cyber are around 800B (GLM 5.3, DeepSeek v4.1 flash) Moreover for a model like DeepSeek v4.1 flash, you need only 500B weights on wafer; remaining 192B are embeddings can be hosted off chip. This brings down wafer need drastically to less than 10!!! Moreover; prefill needs only half the layers in DS architecture that is going to be commonplace in coming days and KV is quite small; so Cerebras can also greatly perform prefill (something they avoid) and do decode with larger micro batches. It will likely be quite efficient with just 10 wafers; and will do well for Cyber, Design and voice use cases, imho.
This post quotes another post: x.com/joefioti/status/2113344525121003579