GPT-6 Astra in OpenCode 1.18.32
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 18m 28sCost $4.51 reported
Generation timelapse
Generation timelapse (the agent's screen during the run, sped up). This is not the finished video. Download video (365 KB).
Review
No verdict yet.
The full review
OK. Finished on its own after 18.5 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, 1,367 characters. Every result for this prompt
Build a Three.js scene of a huge glowing jellyfish walking through a rainy city at night on very tall, thin, jointed legs, presented as a live disaster news broadcast. The jellyfish's translucent bell pulses with soft bioluminescent light, its tentacles sway beneath it, and each slow stride steps between the towers while rain streaks through the light and wet streets reflect neon signs and headlights. Frame it like a TV emergency broadcast: a main camera feed plus several smaller feeds from other angles (street level, a helicopter, a rooftop) that update as the creature moves; a HUD with a channel logo, a red LIVE badge, a scrolling news ticker, a clock and coordinates; and an event log panel that adds timestamped entries as the jellyfish crosses the city. Keep the mood tense and cinematic, with fog, flickering streetlights and flashing red and blue emergency lights. Everything is built in code from Three.js geometry, shaders and materials; no external models or textures. Build it as index.html (plus any local asset files you create) in the current directory. Three.js may load from a pinned CDN. You may serve it locally and open it in headless chromium (/usr/bin/chromium, software WebGL, no GPU) 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.
How it was made
Step by step
The prompt, every agent turn and tool call (tool name and a one-line summary; inputs and outputs are not shown), then the final message. The full transcript is 4.3 MB; it is published as a scrubbed, compressed download rather than shown on the page.
Show all 74 steps
Loading the timeline…
Download the full transcript OpenCode session export, 114 KB, gzip. Scrubbed: anything that looks like a key, token, e-mail address or phone number is replaced, and account details and the test machine's file paths are removed.
Cost breakdown
Headline cost $4.51, as reported by OpenCode (its own bill for the run). The list-price estimate from the token counts is $4.35.
| Tokens by type | Tokens | $ per 1M | Cost |
|---|---|---|---|
| Input (uncached) | 93 | $10.00 | $0.0009 |
| Cache write | 67,493 | $10.00 | $0.6749 |
| Cache read | 1,296,690 | $1.00 | $1.2967 |
| Output | 47,457 | $50.00 | $2.3729 |
| of which reasoning | 13,956 | billed as output |
- Total computed from tokens
- $4.3454
- Total reported by the harness
- $4.5141
- Difference
- Harness-reported $4.5141 vs list-price estimate $4.3454 (-3.7%); the harness figure is the headline.
OpenCode 1.18.32 reported its own total; token counts x list prices shown beside it. Cache writes priced at the input rate (the price table has no separate cache-write price). The harness reports reasoning tokens separately from output; here output includes them (33,501 visible + 13,956 reasoning), and both bill at the output rate.
Prices: Project price table (vendor list prices), 2026-09-22.
Run notes
- frozen prompt (rewritten): engine/prompts_rewritten/threejs_jellyfish.txt: written by us from the recorded description (zero results)
- plain OpenCode config copy (providers only; no ECC instructions/agents)
- opencode export written to a file (not piped)
- Mode: interactive TUI in tmux (160x48), inside bubblewrap sandbox
- Isolation audit: outside paths seen