NL2GDS: LLM-aided interface for Open Source Chip Design
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915838129340416 |
|---|---|
| author | Eland, Max Thiyagalingam, Jeyan Pamunuwa, Dinesh Weerasekera, Roshan |
| author_facet | Eland, Max Thiyagalingam, Jeyan Pamunuwa, Dinesh Weerasekera, Roshan |
| contents | The growing complexity of hardware design and the widening gap between high-level specifications and register-transfer level (RTL) implementation hinder rapid prototyping and system design. We introduce NL2GDS (Natural Language to Layout), a novel framework that leverages large language models (LLMs) to translate natural language hardware descriptions into synthesizable RTL and complete GDSII layouts via the open-source OpenLane ASIC flow. NL2GDS employs a modular pipeline that captures informal design intent, generates HDL using multiple LLM engines and verifies them, and orchestrates automated synthesis and layout. Evaluations on ISCAS'85 and ISCAS'89 benchmark designs demonstrate up to 36% area reduction, 35% delay reduction, and 70% power savings compared to baseline designs, highlighting its potential to democratize ASIC design and accelerate hardware innovation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_05489 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | NL2GDS: LLM-aided interface for Open Source Chip Design Eland, Max Thiyagalingam, Jeyan Pamunuwa, Dinesh Weerasekera, Roshan Hardware Architecture Computers and Society Logic in Computer Science Systems and Control The growing complexity of hardware design and the widening gap between high-level specifications and register-transfer level (RTL) implementation hinder rapid prototyping and system design. We introduce NL2GDS (Natural Language to Layout), a novel framework that leverages large language models (LLMs) to translate natural language hardware descriptions into synthesizable RTL and complete GDSII layouts via the open-source OpenLane ASIC flow. NL2GDS employs a modular pipeline that captures informal design intent, generates HDL using multiple LLM engines and verifies them, and orchestrates automated synthesis and layout. Evaluations on ISCAS'85 and ISCAS'89 benchmark designs demonstrate up to 36% area reduction, 35% delay reduction, and 70% power savings compared to baseline designs, highlighting its potential to democratize ASIC design and accelerate hardware innovation. |
| title | NL2GDS: LLM-aided interface for Open Source Chip Design |
| topic | Hardware Architecture Computers and Society Logic in Computer Science Systems and Control |
| url | https://arxiv.org/abs/2603.05489 |