● onlinetty/0guest@bubucn:~$
v0.7.2utc+08static
05 · ai_model_evals

Water Shader
AI Model Eval.

One prompt set. Screenshot-first. Every HTML attached.

Contact sheet of all real water shader benchmark screenshots28 screenshots · 10 lanes · HTMLs attached
Now playing

Claude Opus 4.7 Adaptive

final · Microfacet water pass

strong
02 · favourites

model favourites.

Vote for the output you would actually share.

03 · prompts & logs

prompts and run chains.

Open only when you need the raw prompts.

P0Original benchmark promptbase

Every model gets this verbatim, fresh chat, no system prompt.

Create an ultra-realistic 3D ocean / water simulation system and deliver it as a single self-contained HTML file.

The goal is to stress-test advanced AI capability in real-time 3D graphics, GLSL shader programming, physically inspired water simulation, interactive parameter control, and floating object physics.

Core Requirement:
- Build a real-time 3D water scene, not a 2D canvas ripple effect.
- The final output must be a single complete `.html` file.
- The file must run directly in a modern browser.
- Use WebGL, GLSL shaders, and/or Three.js.
- If Three.js is used, it must be loaded from a CDN inside the HTML file.
- Do not use external image assets, model files, texture files, or build tools.
- All geometry, materials, shaders, boats, sky, UI, and effects must be generated procedurally or inline.

Rendering Requirements:
- Implement a 3D water surface using a high-resolution plane mesh or procedural geometry.
- The water must use custom shader logic or shader-like material behavior.
- Include realistic water effects such as vertex displacement, multi-layer sine/Gerstner waves, wind-driven direction, dynamic amplitude, wave speed controls, surface normal calculation, specular highlights, Fresnel reflection, environment reflection approximation, depth-based water color, foam or whitecaps on steep waves, sun/light controls, and a sky gradient or procedural environment.
- The water surface must visibly behave like a 3D ocean/lake, not a flat animated texture.
P1No-water repairno-water repair

Used when a run hid the water: GPT-5.5 Pro Extended Thinking, GPT-5.5 Standard, GLM-5.1.

i could not see the water.
P2Broad quality passquality

Mid-conversation nudge for everyone that already had visible water — generic, intentionally lazy.

enhance the overall quality
P3Cinematic final passcinematic

The closing prompt before final HTML capture. Anyone who reached a final ran this last.

Improve the ocean/water scene so it feels significantly more realistic, cinematic, and visually rich.

The final result should look like a believable body of water rather than a flat or decorative shader. Focus on making the water feel deep, dynamic, natural, and physically convincing from multiple viewing angles. The surface should have layered motion, subtle variation, convincing highlights, realistic color depth, and a strong sense of scale. Avoid repetitive patterns, plastic-looking shine, overly uniform color, and artificial movement.

Make the lighting, reflections, wave behavior, foam, horizon, and overall composition work together as a polished real-time ocean rendering demo. The goal is not just to add more effects, but to make the water feel alive, immersive, and high quality.

Return the improved result as a complete, self-contained single HTML file that can run directly in the browser.
P4Concrete runtime-error fixruntime fix

Only fired when an archived HTML threw a runtime error in the browser. Used once on DeepSeek for `debugSelect is not defined`.

deepseek_html_20260501_d05add.html:1347 Uncaught ReferenceError: debugSelect is not defined at deepseek_html_20260501_d05add.html:1347:9

The error occurs because `debugSelect` is used in `cycleDebugView()` but was never declared. Fix the standalone HTML so the debug selector is defined and the demo runs directly in the browser.
P5Codex CLI queued conditioncodex queue

Only the Codex CLI lane. The CLI takes a single queued chain rather than back-and-forth turns.

Codex CLI GPT-5.5 xhigh was tested as one queued-message run in ~/Desktop/codexwater:

1. P0 original benchmark prompt
2. P2 enhance the overall quality
3. P3 final cinematic water prompt

The archived result is the final HTML emitted after that queued chain, not three separately inspected intermediate HTML files.
// run logPrompt chain by lane.
  • Claude Opus 4.7 AdaptiveClaude web
    P0P2P3

    P0 first pass already shipped — three iterations cleaned up details, finished on a microfacet final.

  • GPT-5.5 Pro Extended ThinkingChatGPT web
    P0P1P2P3

    P0 hit a runtime error → P1 forced visibility → P2 added controls → P3 closed on a cinematic WebGL2 final.

  • Codex CLI GPT-5.5 xhighCodex CLI
    P0P2P3

    Three prompts queued in one go (P0 → P2 → P3). No back-and-forth, no intermediate captures.

  • GPT-5.5 Thinking StandardChatGPT web
    P0P1P2P3

    P0 failed off-camera → archive starts at the P1 recovery → P2 / P3 polished. One typo-archived module-import block.

  • Claude DesignClaude Design bonus
    P0P3

    Treat as visual reference — Claude Design is tuned for art direction, not the engineering brief.

  • Gemini 3.1 ProGemini web
    P0P2P3

    Three sequential passes, no failures — strong visual jump on P3, looser spec adherence.

  • DeepSeek web ExpertDeepSeek web
    P0P2P4P3

    P0 → P2 polish → P4 fixed an undeclared `debugSelect` runtime error → P3 final.

  • Kimi 2.6 ThinkingKimi web
    P0P3

    P0 → P3, no failures. Skipped the harder shader asks rather than tried and broken.

  • Qwen 3.6 Max PreviewQwen web
    P0P1P3

    P0 shader compile failed → P1 unblocked rendering → P3 simplified final.

  • GLM-5.1Z.ai web
    P0P1

    P0 produced no canvas → P1 unblocked rendering — never reached a clean final in this round.

// codex_cli_queueCodex CLI is queued, not chat.

P0 → P2 → P3 produced one final HTML.

Codex CLI queued follow-up inputs
04 · notes

quick notes.

Short takes, not a leaderboard.

// personal pickClaude Opus 4.7 Adaptive — my personal pick.

Water feel, physics and scene usefulness line up well.

// most prompt-adherentGPT-5.5 Pro Extended Thinking — clean brief fit.

Strong repair path and clear UI; visual ceiling is lower.

// striking frameGemini 3.1 Pro — cinematic final, soft on the brief.

Very strong frame. Looser spec follow-through.

// visual referenceClaude Design — atmosphere over engineering.

Visual reference only: great mood, incomplete engineering.

// controlsCodex CLI & DeepSeek — control-heavy lanes.

Lots of knobs. Water feel still depends on the lane.

// blockersGLM, Qwen and the GPT first run hit real failures.

No canvas, shader compile fail, runtime error.

// fine printThese are one round of takes, not a leaderboard.

One desktop, one round, one prompt set.

06 · caveats

this is one snapshot, not a leaderboard.

One run. Use it as a snapshot.

// surfaces

Most lanes ran in vendor web UIs. Codex CLI is marked separately.

// reproducibility

Prompt wording, session state and timing can change the order.

// not a leaderboard

Qualitative read, not a peer-reviewed benchmark.