Layer 2 · head to head
Cerebras WSE-3 vs Groq LPU
Speed against cost. Cerebras is 3-8x faster; Groq is cheaper per million tokens.
| Cerebras WSE-3 | Groq LPU | |
|---|---|---|
| Can you get it | Rentable by the hour from multiple clouds; purchasable at scale. | Rentable by the hour from multiple clouds; purchasable at scale. |
| Memory | 44 GB on-chip SRAM | 230 MB SRAM per chip, no HBM |
| Bandwidth | 21 PB/s on-chip | 80 TB/s on-chip |
| Compute | 900,000 AI cores, 4 trillion transistors | Deterministic inference architecture |
| Power | Not published per unit | Not published per chip |
| Interconnect | On-wafer fabric | Chip-to-chip fabric |
| Software | Cerebras SDK, PyTorch | GroqCloud API, GroqWare |
| Workload | inference | inference |
Cerebras WSE-3
For: Latency-critical inference. Delivered Llama 4 Maverick at 2,500 tokens per second per user — a figure no GPU cluster approaches.
The catch: 44 GB of SRAM is the whole memory system. There is no HBM to fall back on, so the model has to fit the architecture. This is a specialist part, not a general one.
Economics: Roughly 3-8x Groq's raw throughput depending on the benchmark, but Groq is cheaper per million tokens. Speed and cost point at different chips here.
Groq LPU
For: Cheap, fast token generation via API. The lowest cost per million tokens among the specialist inference parts.
The catch: 230 MB per chip means the model is split across many chips — Llama2-70B at Groq's headline speed required 576 chips, roughly eight racks. The per-chip spec is misleading unless you read it as a per-cluster architecture.
Economics: Cheaper per million tokens than Cerebras while being materially slower. If your product is priced per token rather than per millisecond, that is the right trade.
Before either — can you power it?
Not published per unit against Not published per chip. In 2026 that comparison usually matters more than the FLOPS one: the US interconnection queue exceeds 2,600 GW with waits approaching five years, and roughly 80% of projects withdraw before energising.
Want this whole layer as one table?
The accelerators on these pages as a single CSV — memory, bandwidth, compute, power, interconnect, software stack, availability, who each part is for and the catch — ready to sort, filter or load into a notebook.
Get the silicon table — free for research and evaluation
Four short lines and the download starts at once (CC-BY-NC-4.0, built from these pages the moment you ask). What you type is used only to answer the request and is never sold. Prefer Hugging Face? Request it there.
Using it inside a product, a client report or anything you sell? That needs a commercial licence.