SambaNova · RDU · Layer 2
SambaNova SN40L
Serving many large models from one system. The three-tier memory design exists to swap models in and out rather than to hold one resident.
rent Rentable by the hour from multiple clouds; purchasable at scale.
Specifications
- Memory
- 520 MiB SRAM + 64 GiB HBM + up to 1.5 TiB DDR per socket
- Bandwidth
- Over 1 TB/s DDR-to-HBM model streaming
- Compute
- Reconfigurable dataflow
- Power
- Not published
- Interconnect
- Socket-to-socket fabric
- Software
- SambaFlow
- Form factor
- accelerator
- Workload
- inference
Verified 2026-09-09. Fields reading “not published” are exactly that — we do not estimate a figure a vendor withholds.
The catch
The most unusual architecture here, so the least portable. Independently verified at 1,084 tokens/s for Llama3-8B — but on 16 sockets, which is the number that matters.
The economics
Its argument is models-per-dollar rather than tokens-per-dollar. If you serve one model, this is the wrong shape.
Cost per million tokens is the metric that decides most real purchases, and there is no neutral benchmark for it — every published figure comes from a company selling one side of the comparison. Read all of them, ours included, as claims rather than measurements.
The power question underneath this
A rack of frontier accelerators draws 120–200 kW against a 2026 average of about 27 kW, and the US grid interconnection queue exceeds 2,600 GW with waits approaching five years. Whether you can energise this part is now a harder question than whether you can buy it.
Layer 1 — Energy · The interconnection queue · Rack power density · Tokens per watt
Where you actually rent this
Choosing the part is the easy half. The same accelerator rents for wildly different money depending only on who you rent it from — the cheapest and dearest published on-demand price for an H100 sit 5.6× apart — and the provider, not the chip, is what decides whether you can leave later.
Compared against
- SambaNova SN40L vs Groq LPU — Many models on one system against one model served cheaply.Compare with Groq LPU
Related silicon
- Cerebras WSE-3 — 44 GB on-chip SRAM, rent
- Groq LPU — 230 MB SRAM per chip, no HBM, rent
- AMD Helios (Instinct MI455X) — 31 TB of HBM4 per rack across 72 MI455X GPUs; 432 GB of HBM4 per GPU., rent
- AMD Instinct MI300X — 192 GB HBM3, rent
- AMD Instinct MI350X — 288 GB HBM3E, rent
- AMD Instinct MI355X — 288 GB HBM3E, rent
Sources
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.