Cerebras WSE-3
Single-wafer accelerators delivering the fastest published open-model token rates.
Full Cerebras WSE-3 record →Head to head
Cerebras WSE-3 vs AMD Instinct MI355X
Cerebras WSE-3 (Cerebras) and AMD Instinct MI355X (AMD) both sit in Chips & Compute. Cerebras WSE-3 shipped more tracked changes in the last 30 days (9 vs 3). Every value below comes from the latest crawl of the vendors' own pages.
| Attribute | Cerebras WSE-3 Cerebras | AMD Instinct MI355X AMD |
|---|---|---|
| Segment | Chips & Compute | Chips & Compute |
| Version | CS-4 | MI355X |
| Entry cost | Free trial credits available; developer tier has free promotional credits. | Enterprise hardware, pricing not published |
| Pricing tiers | Developer Tier / Free Trial free | Open Source free |
| Model stack | Codex-Spark · Gemma · Gemma-4-31B · GLM-4.7 · GPT 5.6 Sol · GPT-5.6-Sol-Ultrafast · Kimi K2.6 · Kimi K3.1 · Llama · Llama / Qwen / GPT-OSS · Qwen · QWEN 3.8 27B · WSE-3 | CDNA · CDNA 4 · ROCm |
| Context window | — | — |
| Public API | ||
| Routes models | ||
| Changes / 30d | 9 | 3 |
| Origin | Global | Global |
| Capabilities | wafer-scale engine, record token/s on open models, CS-3 systems, training clusters, Wafer-scale AI inference, Cerebras PyTorch, ModelZoo, Low-level SDK | 288GB HBM3E, FP4/FP6 support, ROCm 7, open rack designs, DPX compute partition mode, CPX compute partition mode, NPS2 memory partitioning, GPU virtualization support |
AMD Instinct MI355X
288GB HBM3E accelerators with the ROCm software stack, used by several inference providers.
Full AMD Instinct MI355X record →