Lightweight edge chip
Forge
A complete language model inside a sub-watt handheld device.
Lightweight edge chip
17,000token/s
MaskROM compute-in-memory: Stable models, high volume, lowest power and marginal cost.
Specifications
- Core model
- Qwen 3.5 0.8B
- Decode speed
- 17,000 token/s
- System power
- Sub-watt class
- Route
- MaskROM compute-in-memory
- 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
- Handheld translators
- Voice memo hardware
- Offline in-car voice assistants
Ask about Forge
Send the target model, performance and power goals, interfaces and timeline. We reply with a fit assessment and a sample plan.