Spaces:
Running
Running
| """HTML content for the TuRTLe leaderboard.""" | |
| HEADER_HTML = """ | |
| <style> | |
| .turtle-logo-container { | |
| padding: 0.5rem 0 4rem 0; | |
| margin-bottom: -6rem; | |
| position: relative; | |
| z-index: 0; | |
| overflow: visible; | |
| } | |
| .turtle-logo { | |
| width: 17rem; | |
| padding: 0; | |
| transition: all 300ms; | |
| } | |
| @media (prefers-color-scheme: dark) { | |
| .turtle-logo { | |
| filter: drop-shadow(0 0 2em #026e00); | |
| } | |
| .turtle-logo:hover { | |
| filter: drop-shadow(0 0 0em #026e00); | |
| } | |
| } | |
| </style> | |
| <div align="center" class="turtle-logo-container"> | |
| <picture> | |
| <source srcset='/gradio_api/file=turtle_dark.svg' media='(prefers-color-scheme: dark)'> | |
| <source srcset='/gradio_api/file=turtle_light.svg' media='(prefers-color-scheme: light)'> | |
| <img src='/gradio_api/file=turtle_dark.svg' alt='TuRTLe Logo' class='turtle-logo'/> | |
| </picture> | |
| </div> | |
| """ | |
| NAV_BUTTONS_HTML = """ | |
| <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.0.0/css/all.min.css"> | |
| <script defer src="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.0.0/js/all.min.js"></script> | |
| <div style="text-align: center; margin-bottom: 0px; margin-top: 0px;"> | |
| <a href="https://github.com/HPAI-BSC/TuRTLe" target="_blank" style="text-decoration: none; margin-right: 10px;"> | |
| <button style="background: #333; color: white; padding: 10px 14px; border-radius: 8px; border: none; font-size: 16px; cursor: pointer;"> | |
| GitHub Repo | |
| </button> | |
| </a> | |
| <a href="http://arxiv.org/abs/2504.01986" target="_blank" style="text-decoration: none; margin-right: 10px;"> | |
| <button style="background: #b31b1b; color: white; padding: 10px 14px; border-radius: 8px; border: none; font-size: 16px; cursor: pointer;"> | |
| arXiv MLCAD 2025 | |
| </button> | |
| </a> | |
| <a href="mailto:[email protected]?subject=TuRTLe%20leaderboard%20new%20entry&body=Link%20to%20HuggingFace%20Model:" style="text-decoration: none;"> | |
| <button style="background: #00674F; color: white; padding: 10px 14px; border-radius: 8px; border: none; font-size: 16px; cursor: pointer;"> | |
| How to submit | |
| </button> | |
| </a> | |
| <p style="margin-top: 15px;">If you have any inquiries or wish to collaborate: | |
| <a href="mailto:[email protected]">[email protected]</a> | |
| </p> | |
| </div> | |
| """ | |
| INTRO_HTML = """ | |
| <div style=" margin-top:-10px !important;"> | |
| <p style="margin-bottom: 15px; text-align: start !important;"> | |
| Welcome to the TuRTLe Model Leaderboard! TuRTLe is a | |
| <b>unified evaluation framework designed to systematically assess Large Language Models (LLMs) in RTL (Register-Transfer Level) generation</b> | |
| for hardware design. | |
| Evaluation criteria include <b>syntax correctness, functional accuracy, synthesizability, and post-synthesis quality</b> | |
| (PPA: Power, Performance, Area). TuRTLe integrates multiple benchmarks to highlight strengths and weaknesses of available LLMs. | |
| Use the filters below to explore different RTL benchmarks, simulators and models. | |
| </p> | |
| <div class="change-log"> | |
| <p style="margin-top:10px; text-align:start !important;"> | |
| <span style="font-variant:small-caps; font-weight:bold;">UPDATE (DEC 2025):</span> We introduce a brand new benchmark for Module Completion, based on <a href="https://tinytapeout.com/" target="_blank">Tiny Tapeout</a> with more than 1000 real-world designs | |
| </p> | |
| <p style="margin-top:-6px; text-align:start !important;"> | |
| <span style="font-variant:small-caps; font-weight:bold;">UPDATE (NOV 2025):</span> We release a new codebase <a href="https://github.com/HPAI-BSC/TuRTLe" target="_blank">TuRTLe v2</a> with API support and local Docker evaluation. Added <span>Kimi K2 Inst</span>, <span>DeepSeek V3.1 Terminus</span>, and <span>Google's Gemini 2.5 Flash</span> | |
| </p> | |
| <p style="margin-top:-6px; text-align:start !important;"> | |
| <span style="font-variant:small-caps; font-weight:bold;">UPDATE (OCT 2025):</span> Added <span>Hermes-4-14B</span>, <span>Qwen3-8B</span>, and <span>Seed-OSS-36B</span> to the leaderboard. Implemented Other Models tab and moved models to it | |
| </p> | |
| <p style="margin-top:-6px; text-align:start !important;"> | |
| <span style="font-variant:small-caps; font-weight:bold;">UPDATE (SEPT 2025):</span> Added <span>gpt-oss-20b</span> and <span>gpt-oss-120b</span> to the leaderboard | |
| </p> | |
| <p style="margin-top:-6px; text-align:start !important;"> | |
| <span style="font-variant:small-caps; font-weight:bold;">UPDATE (JULY 2025):</span> Our TuRTLe paper was accepted to | |
| <a href="https://mlcad.org/symposium/2025/" target="_blank">MLCAD 2025</a> in September (Santa Cruz, CA) | |
| </p> | |
| <p style="margin-top: -6px; text-align: start !important;"> | |
| <span style="font-variant: small-caps; font-weight: bold;">UPDATE (JUNE 2025):</span> We make our framework open-source on GitHub and we add 7 new recent models! For a total of 40 base and instruct models and 5 RTL benchmarks | |
| </p> | |
| <svg fill="#000000" height="800px" width="800px" version="1.1" id="Capa_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" | |
| viewBox="0 0 27.963 27.963" xml:space="preserve"> | |
| <g> | |
| <g id="c129_exclamation"> | |
| <path d="M13.983,0C6.261,0,0.001,6.259,0.001,13.979c0,7.724,6.26,13.984,13.982,13.984s13.98-6.261,13.98-13.984 | |
| C27.963,6.259,21.705,0,13.983,0z M13.983,26.531c-6.933,0-12.55-5.62-12.55-12.553c0-6.93,5.617-12.548,12.55-12.548 | |
| c6.931,0,12.549,5.618,12.549,12.548C26.531,20.911,20.913,26.531,13.983,26.531z"/> | |
| <polygon points="15.579,17.158 16.191,4.579 11.804,4.579 12.414,17.158 "/> | |
| <path d="M13.998,18.546c-1.471,0-2.5,1.029-2.5,2.526c0,1.443,0.999,2.528,2.444,2.528h0.056c1.499,0,2.469-1.085,2.469-2.528 | |
| C16.441,19.575,15.468,18.546,13.998,18.546z"/> | |
| </g> | |
| <g id="Capa_1_207_"> | |
| </g> | |
| </g> | |
| </svg> | |
| </div> | |
| </div> | |
| """ | |
| LC_FOOTNOTE_HTML = """ | |
| <div id="lc-footnote" style="font-size: 13px; opacity: 0.6; margin-top: -5px; z-index:999; text-align: left;"> | |
| <span style="font-weight: 600; opacity: 1;">*</span> | |
| <em>Line Completion</em> excludes "reasoning" models since this task targets quick auto-completion<br/> | |
| </div> | |
| """ | |
| ABOUT_US_HTML = """ | |
| <div style="max-width: 800px; margin: auto; padding: 20px; border: 1px solid #ccc; border-radius: 10px;"> | |
| <div style="display: flex; justify-content: center; align-items: center; gap: 5%; margin-bottom: 20px;"> | |
| <img src='/gradio_api/file=hpai_logo_grad.png' alt='HPAI Group Logo' style="width: 45%;"/> | |
| <img src='/gradio_api/file=bsc-logo.png' alt='BSC Logo' style="width: 25%;"/> | |
| </div> | |
| <p style="font-size: 16px; text-align: start;"> | |
| The <b>High-Performance Artificial Intelligence (HPAI)</b> group is part of the | |
| <a href="https://bsc.es/" target="_blank">Barcelona Supercomputing Center (BSC)</a>. | |
| This leaderboard is maintained by HPAI as part of our commitment to <b>open science</b>. | |
| </p> | |
| <ul style="font-size: 16px; margin-bottom: 20px; margin-top: 20px;"> | |
| <li><a href="https://hpai.bsc.es/" target="_blank">HPAI Website</a></li> | |
| <li><a href="https://github.com/HPAI-BSC/" target="_blank">HPAI GitHub Organization Page</a></li> | |
| <li><a href="https://huggingface.co/HPAI-BSC/" target="_blank">HPAI Hugging Face Organization Page</a></li> | |
| </ul> | |
| <p style="font-size: 16px; margin-top: 15px;"> | |
| Feel free to contact us: | |
| </p> | |
| <p style="font-size: 16px;">Email: <a href="mailto:[email protected]"><b>[email protected]</b></a></p> | |
| </div> | |
| """ | |
| REFERENCES_HTML = """ | |
| <div style="max-width: 800px; margin: auto; padding: 20px; border: 1px solid #ccc; border-radius: 10px;"> | |
| <ul style="font-size: 16px; margin-bottom: 20px; margin-top: 20px;"> | |
| <li>Claire Wolf and the YosysHQ contributors. Yosys Open SYnthesis Suite. <a href="https://yosyshq.net/yosys/">https://yosyshq.net/yosys/</a></li> | |
| <li><a href="https://github.com/bigcode-project/bigcode-evaluation-harness" target="_blank">Code Generation LM Evaluation Harness</a> by the BigCode project</li> | |
| <li>Williams, S. Icarus Verilog [Computer software]. <a href="https://github.com/steveicarus/iverilog" target="_blank">https://github.com/steveicarus/iverilog</a></li> | |
| <li>Snyder, W., Wasson, P., Galbi, D., & et al. Verilator [Computer software]. <a href="https://github.com/verilator/verilator" target="_blank">https://github.com/verilator/verilator</a></li> | |
| <li>RTL-Repo: Allam and M. Shalan, "Rtl-repo: A benchmark for evaluating llms on large-scale rtl design projects," in 2024 IEEE LLM Aided Design Workshop (LAD). IEEE, 2024, pp. 1–5.</li> | |
| <li>VeriGen: S. Thakur, B. Ahmad, H. Pearce, B. Tan, B. Dolan-Gavitt, R. Karri, and S. Garg, "Verigen: A large language model for verilog code generation," ACM Transactions on Design Automation of Electronic Systems, vol. 29, no. 3, pp. 1–31, 2024. </li> | |
| <li>VerilogEval (I): M. Liu, N. Pinckney, B. Khailany, and H. Ren, "Verilogeval: Evaluating large language models for verilog code generation," in 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD). IEEE, 2023, pp. 1–8.</li> | |
| <li>VerilogEval (II): N. Pinckney, C. Batten, M. Liu, H. Ren, and B. Khailany, "Revisiting VerilogEval: A Year of Improvements in Large-Language Models for Hardware Code Generation," ACM Trans. Des. Autom. Electron. Syst., feb 2025. https://doi.org/10.1145/3718088</li> | |
| <li>RTLLM: Y. Lu, S. Liu, Q. Zhang, and Z. Xie, "Rtllm: An open-source benchmark for design rtl generation with large language model," in 2024 29th Asia and South Pacific Design Automation Conference (ASP-DAC). IEEE, 2024, pp. 722–727.</li> | |
| </ul> | |
| </div> | |
| """ | |
| OTHER_MODELS_HTML = """ | |
| <div style="max-width: 800px; margin: auto; padding: 20px; border: 1px solid #ccc; border-radius: 10px;"> | |
| <p style="font-size: 16px; text-align: start;"> | |
| These models were previously listed on the main leaderboard, evaluated with a potentially deprecated version of TuRTLe, and will no longer be updated. | |
| </p> | |
| </div> | |
| """ | |