GPT-6 Sol in Codex CLI 0.156.1
skills off
Zoom, inline player and more ways to view
The model's own page, run in a sandboxed frame only when you ask (a page that needs its own origin opens in a new tab on a separate demo site). Phones get a full-screen player capped at 1.5x pixel density (2x on larger screens) so it stays smooth; the original page runs uncapped. It starts in its auto-tour (T switches to play mode on a keyboard).
Wall time 6m 34sCost $0.86 estimated
Review
No verdict yet.
The full review
OK. Finished on its own after 6.6 min.
Scores
AI judge
Not scored
No rubric for this prompt
Visitor votes and reviewer grades are not available yet.
- AI judge
- AI judge: the test runner (model claude-opus-5-5), scoring by visual review of the 1600 px render against the pre-registered rubric. No AI-judge score exists for this result: there is no rubric for this prompt, so no score is shown and none is made up.
- Visitors
- Public voting with a blind reveal is coming: you score the output first, then see the other scores.
- Reviewer
- The reviewer's own grade appears here once it is added.
Scores compare results for the same prompt only. AI-judge boards rank runs within one prompt, equal totals sharing a rank; scores from different prompts are never averaged.
Downloads
Files actually published for this run. Original output, site-made previews and post-run conversions are labelled separately.
The exact prompt
as run, 703 characters. Every result for this prompt
In one prompt, build a playable 3D kart racer in the browser with Three.js: one closed track with banked turns, 3 AI opponents, lap counter, 3 laps, finish screen with time. Keyboard controls. All assets procedural (no external files). Single index.html plus JS. Save the main page as index.html in the current directory (plus any local asset files you create), with no build step. You may serve it locally and open it in headless chromium (/usr/bin/chromium, software WebGL, no GPU, no WebGPU) to screenshot and inspect your own output, and iterate on your own before finishing. Close any browser you open as soon as you've taken your screenshot. Finish with a short README.md describing the controls.
How it was made
Versions saved during the run
3 intermediate outputs, in the order the model wrote them.
No transcript was delivered with this result.
Cost breakdown
Headline cost $0.86, estimated from the token counts at list price.
| Tokens by type | Tokens | $ per 1M | Cost |
|---|---|---|---|
| Input (uncached) | 60 | $2.00 | $0.0001 |
| Cache write | 70,664 | $2.00 | $0.1413 |
| Cache read | 956,767 | $0.20 | $0.1914 |
| Output | 52,361 | $10.00 | $0.5236 |
| of which reasoning | 27,009 | billed as output |
- Total computed from tokens
- $0.8564
- Total reported by the harness
- none
Computed from harness-reported token counts x list prices. Cache writes priced at the input rate (the price table has no separate cache-write price). No request above 272,000 input tokens (20 checked): standard rates.
Prices: Project price table, 2026-09-25.
Run notes
- frozen prompt (written): X-idea prompt queue (frozen in prompts_queue), MR wave 4 Three.js/WebGPU expansion; category threejs; engine threejs harness suffix appended; zero results
- plain CODEX_HOME copy (config.toml only, run work/ trusted; no AGENTS.md/skills/state)
- Mode: interactive TUI in tmux (160x48), inside bubblewrap sandbox
- Isolation audit: outside paths seen
- Renderer: Three.js WebGLRenderer (site/index.html line 213; bundled three.module.js imported at line 186). No WebGPU needed: the live demo runs in any WebGL browser.
- Kart pair: static scene cleared only; playability and motion on HOLD; no videos. No quality winner is claimed between the two kart runs: their configs differ (Claude Code, Opus 5.5, effort high vs Codex CLI, GPT-6 Sol, effort max) and both are unscored.
- Poster: captured from the served demo in headless Chromium (software WebGL, no GPU) about 8 s after starting a race; the run's own render.png shows the title screen.