Flagship heavy-load chip
Trident
Run a 35B model locally at 15,000 token/s with a 70 W system budget.
Flagship heavy-load chip
15,000token/s
Near-memory compute: Large, fast-evolving models in heavy-load nodes.
Specifications
- Core model
- Up to Qwen 3.5 35B
- Decode speed
- 15,000 token/s
- System power
- 70 W design target
- Route
- Near-memory compute
- Process
- SMIC 28 nm
- Package / interfaces
- Per project
- Development kit
- FPGA evaluation board + model mapping tools
Figures on this page are current product definitions and design targets. Final specifications, test conditions, availability and delivery versions are confirmed per project.
Typical applications
- Embodied robots
- Industrial edge nodes
- Air-gapped terminals
Ask about Trident
Send the target model, performance and power goals, interfaces and timeline. We reply with a fit assessment and a sample plan.