← All hardware
Reconfigurable Dataflow Unit with three-tier memory for trillion-param AI
Pros
- Excellent for large MoE inference
- Fast token generation
- Turnkey full-stack offering
- Intel-backed funding in 2026
Cons
- Proprietary dataflow stack
- Appliance/cloud model, not loose chips
- Niche vs GPU mainstream
- Smaller ecosystem
✓ Where it shines / best for
- Fast, high-throughput LLM inference for enterprises and developers
- Organizations needing on-prem/private model deployment
- Agentic and multi-model serving with model switching
✕ Not the best fit for
- Teams committed to the NVIDIA CUDA software ecosystem
- Edge or on-device inference
- Buyers wanting transparent public hardware list pricing
Features
- ✓ AI inference
- ✓ API access
- ✓ High Throughput
- ✓ Free tier
- ✓ Dataflow Architecture
- ✓ LLM Inference
- ✓ Model Switching
- ✓ On Premise
- ✓ Rdu
Pricing
| Plan | Price | Billing | Notes |
|---|---|---|---|
| SambaNova Cloud - Free | $0 | free | Free developer tier with API access and rate limits on hosted open models |
| SambaNova Cloud - Developer (pay-as-you-go) | Usage-based per token | per token | Per-million input/output token pricing; varies by model (Llama, DeepSeek, Qwen, etc.) |
| SambaNova Cloud - Enterprise | Custom | custom | Dedicated capacity, higher rate limits, SLAs; contact sales |
| DataScale / on-prem (SN40L systems) | Custom | one-time | Full-stack rack hardware sold by quote; pricing not public |
Pricing verified from the official source. Prices change often — confirm on the vendor's site before buying.
Specifications
| use | Large-model inference and training |
| power | Data center scale |
| memory | 520MB SRAM + HBM + DRAM |
| performance | 640 BF16 TFLOPs |
| architecture | RDU SN40L (TSMC 5nm, 2.5D) |
Sponsored
A full review is being generated for this product and will appear here shortly.