Kimi K3 in OpenCode 1.18.35
skills on · OpenCode 1.18.35 + skills on (54: 33 video + 21 web/Three.js)Kit: 33 video + 21 web/Three.js · 1 of 54 installed skills used in the session
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 45m 01sCost $2.81 reported
Review
No verdict yet.
The full review
Stopped by us. Stopped by us (box row cap reached (2700s); stopped by the box lane so the run ships as partial); not a model failure.
Scores
Not scored: only runs that finished on their own are scored.
Downloads
Files actually published for this run. Original output, site-made previews and post-run conversions are labelled separately.
The exact prompt
as run, 721 characters. Every result for this prompt
ok here it is - my first Opus 5.5 vibecoded threejs 3D experience my God!! Opus is creative!! I made a road that forgets which way is down. It folds 90° up a yellow wall, runs upside down across a ceiling, and turns into sheet music: steer through the notes to play each street's melody. Watercolour. Ink. A radio on the roof. #vibejam 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
Versions saved during the run
12 intermediate outputs, in the order the model wrote them.
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 32.9 MB; it is published as a scrubbed, compressed download rather than shown on the page.
Show all 118 steps
Loading the timeline…
Download the full transcript OpenCode session export, 118 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 $2.81, as reported by OpenCode (its own bill for the run). The list-price estimate from the token counts is $2.81.
| Tokens by type | Tokens | $ per 1M | Cost |
|---|---|---|---|
| Input (uncached) | 195,559 | $3.00 | $0.5867 |
| Cache write | 0 | $3.00 | $0 |
| Cache read | 3,574,174 | $0.30 | $1.0723 |
| Output | 77,013 | $15.00 | $1.1552 |
| of which reasoning | 51,072 | billed as output |
- Total computed from tokens
- $2.8141
- Total reported by the harness
- $2.8141
- Difference
- matches
OpenCode 1.18.35 reported its own total; token counts x list prices shown beside it. The harness reports reasoning tokens separately from output; here output includes them (25,941 visible + 51,072 reasoning), and both bill at the output rate.
Prices: https://platform.kimi.ai/docs/pricing/chat, 2026-09-24.
Run notes
- frozen prompt (reconstructed): core requirements sha256 7eb2c54fac50, full harness prompt sha256 7eb2c54fac50; exact prompt text used by the existing harness rows of this test on the runs branch (copied from folding_road_threejs__claude-code__fable-5-1__attempt1/meta.json, sha256 verified); Beta half of the MR
- no TUI: the harness ran in its non-interactive mode, so there are no TUI screenshots; the timelapse is built from version renders only
- OpenCode config (providers only, permission allow-all except question; no agents/instructions) + both default skill kits in xdg/config/opencode/skill (skills=on)
- infra retries: 0
- in the published transcript the sandbox HOME directory is shown as <HOME>
- skills=on: 54 skills installed in the harness skill directory before launch (video-33 remotion-dev/skills + heygen-com/hyperframes@0b5db9da + f; web-threejs-21 [redacted]/ai-AI Model Reviewer skills/ (alpha/mr-homebar)@7521074d98c9).
- Mode: headless run mode (opencode run --format json), inside bubblewrap sandbox
- Isolation audit: none
- Renderer: three.js r160 WebGL (ES module via an importmap) from a local vendor/three.module.js shipped with the page, so nothing is loaded from a CDN; canvas 2D is used only to paint textures.
- Card video: the run's own video.mp4 (H.264 yuv420p, unchanged). Poster: the run's own render.png.
- Status: stopped by us. Stopped at the row cap (45 min) before the harness declared it finished; the page shown is what it had built by then.
- Skills on: the run had 54 skills installed (33 video + 21 web/Three.js), as its meta records.
- Kimi K3 served via Fireworks (OpenCode provider fireworks-ai).
- Same frozen prompt as the other rows on this test; configurations differ and all rows are unscored, so this is not a head-to-head result.