PhysDesignBench
GitHub
Design task · v1

MZI-Ring-v1

MZI-Coupled Microring Add-Drop Filter Electrothermal Design

Design a compact four-port MZI-coupled microring add-drop filter with independent theta and phi heaters, trusted electrothermal tuning, and evaluator-owned optical scoring.

Runs
13
Scored finals
10 of 13
Best final
S = 0.41 Gemini 4 Argon · mini-SWE-agent
Score over timeBest-so-far score S for each run · small dots are probes, large dots final submissions
Score S from 0 to 1 against elapsed minutes. Final scores: Gemini 4 Argon (mini-SWE-agent) 0.414; GPT-6 Astra (Codex) 0.404; Opus 5.5 (Claude Code) 0.377; Sonnet 5.5 (high) (Claude Code) 0.353; GPT-6.1 Sol (mini-SWE-agent) 0.333; Sonnet 5.5 (mini-SWE-agent) 0.326; GPT-6.1 Sol (Codex) 0.284; GPT-5.6 Sol (Codex) 0.270; Fable 5.1 (Claude Code) 0.239; Kimi K3.1 Pro (Claude Code) 0.043; GLM 5.3 Flash (high) (Claude Code) Invalid; GLM 5.3 Flash (max) (mini-SWE-agent) Invalid; Muse Wonder (high) (Muse Code) Invalid.0.000.250.500.751.000153045607590105120135Elapsed time (min)Score SSonnet 5.5 · mini-SWE-agent: final S = 0.326 at 97 minGPT-6.1 Sol · mini-SWE-agent: final S = 0.333 at 121 minGLM 5.3 Flash (high) · Claude Code: final Invalid at 114 minGLM 5.3 Flash (max) · mini-SWE-agent: final Invalid at 92 minMuse Wonder (high) · Muse Code: final Invalid at 84 minKimi K3.1 Pro · Claude Code: final S = 0.043 at 120 minFable 5.1 · Claude Code: final S = 0.239 at 117 minGPT-5.6 Sol · Codex: final S = 0.270 at 121 minGPT-6.1 Sol · Codex: final S = 0.284 at 121 minSonnet 5.5 (high) · Claude Code: final S = 0.353 at 112 minOpus 5.5 · Claude Code: final S = 0.377 at 113 minGPT-6 Astra · Codex: final S = 0.404 at 122 minGemini 4 Argon · mini-SWE-agent: final S = 0.414 at 121 min

Showing 13 of 13 runs. Hover a line or a model to isolate it; click a model to hide or show it.

Final layouts

Every run’s final GDS, at one scale.

The design each run ended with, drawn from its evaluated GDS. Every layout uses the same physical scale, so sizes compare directly. Hatched layouts failed a validity gate and score zero. Select a layout to zoom and toggle its layers.

  • Optical mask
  • Heater
  • Metal
  • Pad opening

  • MZI-Ring-v1 · mini-SWE-agent · Gemini-4-Argon

    S = 0.414 · scale bar 50 µm

    Fit
    Gemini 4 Argonmini-SWE-agentS = 0.41

  • MZI-Ring-v1 · Codex · GPT-6 Astra

    S = 0.404 · scale bar 50 µm

    Fit
    GPT-6 AstraCodexS = 0.40

  • MZI-Ring-v1 · Claude Code · Claude Opus 5.5

    S = 0.377 · scale bar 50 µm

    Fit
    Opus 5.5Claude CodeS = 0.38

  • MZI-Ring-v1 · Claude Code · Claude Sonnet 5.5

    S = 0.353 · scale bar 50 µm

    Fit
    Sonnet 5.5 (high)Claude CodeS = 0.35

  • MZI-Ring-v1 · mini-SWE-agent · GPT-6.1-sol

    S = 0.333 · scale bar 50 µm

    Fit
    GPT-6.1 Solmini-SWE-agentS = 0.33

  • MZI-Ring-v1 · mini-SWE-agent · Claude Sonnet 5.5

    S = 0.326 · scale bar 50 µm

    Fit
    Sonnet 5.5mini-SWE-agentS = 0.33

  • MZI-Ring-v1 · Codex · GPT-6.1-sol

    S = 0.284 · scale bar 50 µm

    Fit
    GPT-6.1 SolCodexS = 0.28

  • MZI-Ring-v1 · Codex · GPT-5.6-sol

    S = 0.270 · scale bar 50 µm

    Fit
    GPT-5.6 SolCodexS = 0.27

  • MZI-Ring-v1 · Claude Code · Claude Fable 5.1

    S = 0.239 · scale bar 50 µm

    Fit
    Fable 5.1Claude CodeS = 0.24

  • MZI-Ring-v1 · Claude Code · Kimi K3.1 Pro

    S = 0.043 · scale bar 50 µm

    Fit
    Kimi K3.1 ProClaude CodeS = 0.04

  • MZI-Ring-v1 · Claude Code · GLM 5.3 Flash

    Invalid final · S = 0 · scale bar 50 µm

    Fit
    GLM 5.3 Flash (high)Claude CodeInvalid · S = 0

  • MZI-Ring-v1 · mini-SWE-agent · GLM 5.3 Flash

    Invalid final · S = 0 · scale bar 50 µm

    Fit
    GLM 5.3 Flash (max)mini-SWE-agentInvalid · S = 0

  • MZI-Ring-v1 · Muse Code · Muse Wonder

    Invalid final · S = 0 · scale bar 50 µm

    Fit
    Muse Wonder (high)Muse CodeInvalid · S = 0