TuRTLe - Icarus EM: leaderboard
The "Icarus EM" subtask of TuRTLe. A task from TuRTLe, evaluating LLMs on RTL / Verilog hardware-description-language code generation and completion, checked with hardware simulators.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around April 2027. 25 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 Coder 480B A35B | 38.88 | #279 |
| 2 | Qwen 2.5 72B | 37.44 | #499 |
| 3 | Llama 3.1 405B | 34.62 | #466 |
| 4 | Kimi K2 Instruct (0905) | 33.65 | #257 |
| 5 | Qwen 2.5 32B | 28.93 | #546 |
| 6 | Hermes-4-14B | 28.36 | #751 |
| 7 | starchat2-15B-v0.1 | 13.42 | #832 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=turtle-icarus-em · How It Works · Data refreshed daily, snapshot 2026-10-20.