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 AWS Trainium2
Cerebras WSE-3 (Cerebras) and AWS Trainium2 (AWS) both sit in Chips & Compute. Cerebras WSE-3 shipped more tracked changes in the last 30 days (10 vs 5). Every value below comes from the latest crawl of the vendors' own pages.
| Attribute | Cerebras WSE-3 Cerebras | AWS Trainium2 AWS |
|---|---|---|
| Segment | Chips & Compute | Chips & Compute |
| Version | WSE-3 | AWS Trainium2 |
| Entry cost | Free tier / developer tier pricing available (contact/partner integrations) | Pricing not published on the provided pages (contact AWS or check EC2 pricing page for details) |
| Pricing tiers | Developer Tier free | — |
| Model stack | Codex-Spark · Gemma · Gemma-4-31B · GLM-4.7 · GPT 5.6 Sol · GPT-5.6-Sol-Ultrafast · Kimi K2.6 · Llama · Llama / Qwen / GPT-OSS · Qwen · WSE-3 | Neuron compiler · Trainium2 |
| Context window | — | — |
| Public API | ||
| Routes models | ||
| Changes / 30d | 10 | 5 |
| 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 | Trn2 UltraServer, 64-chip NeuronLink domain, Neuron SDK, Rainier cluster, Generative AI training, Generative AI inference, Up to 20.8 FP8 petaflops, 1.5 TB HBM3 memory |
AWS Trainium2
Trn2 UltraServers and Project Rainier capacity, programmed through the Neuron SDK.
Full AWS Trainium2 record →