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. Tap inside for sound.
Wall time 23m 52sCost $4.30 estimated
Review
No verdict yet.
The full review
OK. Finished on its own after 23.9 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, 4,483 characters. Every result for this prompt
Build an interactive e-commerce conversion funnel dashboard from the exact dataset below (CSV: one row per week and acquisition channel with sessions, product views, add-to-cart events, checkouts started, orders and revenue in US dollars). Embed the data in the page verbatim and compute every number from it; do not invent or alter any value. week,channel,sessions,product_views,add_to_cart,checkout_started,orders,revenue_usd W01,Organic search,41078,23808,3022,1664,1007,61604.87 W01,Paid social,29553,16291,2453,1413,439,28577.0 W01,Email,9676,6212,765,421,263,17543.38 W01,Affiliates,12459,8376,1123,612,294,19534.46 W02,Organic search,39462,27334,3685,1940,1028,71414.28 W02,Paid social,30938,17683,2757,1524,410,27158.28 W02,Email,9600,6541,909,539,348,21743.58 W02,Affiliates,12881,8218,1003,504,263,17113.4 W03,Organic search,44423,24608,3329,1856,1259,86784.87 W03,Paid social,30858,18201,2214,1150,460,31397.77 W03,Email,9521,5537,625,323,201,12787.56 W03,Affiliates,12450,7220,807,369,252,14706.22 W04,Organic search,42148,27564,4059,1961,1211,80712.33 W04,Paid social,31643,21078,2811,1606,424,28675.25 W04,Email,8990,5420,621,300,186,12825.63 W04,Affiliates,11171,6965,891,459,260,18220.72 W05,Organic search,40552,24076,2693,1602,1065,73137.37 W05,Paid social,27957,19064,2258,1166,380,22865.57 W05,Email,8419,5117,768,458,333,23433.12 W05,Affiliates,13002,8405,991,519,305,21023.72 W06,Organic search,41353,25171,2806,1464,1037,69935.08 W06,Paid social,31261,20635,2962,1610,478,34375.73 W06,Email,8785,5180,602,350,243,14675.03 W06,Affiliates,10876,6911,846,419,217,13958.06 W07,Organic search,45923,26492,2996,1453,1009,63590.97 W07,Paid social,30650,19483,2477,545,397,24507.5 W07,Email,9762,5572,885,427,274,19092.88 W07,Affiliates,12569,8516,1339,607,290,19871.0 W08,Organic search,39257,22508,3067,1608,1090,80304.82 W08,Paid social,27396,18815,2887,1521,370,24924.69 W08,Email,8732,4905,694,357,256,16113.52 W08,Affiliates,12974,7423,871,490,300,18588.38 W09,Organic search,43769,27584,3167,1885,1164,80776.52 W09,Paid social,32979,21507,2954,1761,474,29511.87 W09,Email,9766,5958,740,349,222,13038.74 W09,Affiliates,11969,7534,986,487,256,15410.29 W10,Organic search,62108,38464,5685,2645,1591,112152.09 W10,Paid social,43357,28900,4203,2065,628,42320.58 W10,Email,9833,5660,688,400,279,19964.74 W10,Affiliates,15861,10838,1246,663,324,20077.73 W11,Organic search,61607,38179,4958,2802,1692,112147.42 W11,Paid social,36623,25030,3720,1756,497,33026.25 W11,Email,9552,5336,669,369,239,14580.06 W11,Affiliates,15696,10500,1268,676,317,19178.7 W12,Organic search,40329,25455,3443,1880,1081,70528.56 W12,Paid social,29585,20344,2424,1159,401,24455.8 W12,Email,8312,5695,677,330,203,12879.42 W12,Affiliates,13056,7443,910,502,296,17318.13 The dashboard needs: an animated funnel chart (sessions to orders) with the drop-off at every stage as a count and a percentage; KPI cards for revenue, orders, conversion rate, average order value and revenue per session, each with its change versus the previous four weeks; a channel comparison of conversion rate and revenue per session; a weekly trend chart with a metric switcher; a stage-by-week heatmap of step conversion rates that makes anomalies obvious, with any week where a step rate falls below 60 percent of its channel average flagged and explained; a what-if panel with sliders to lift each stage's conversion and see projected orders and revenue; filters by channel and week range that update everything; and hover or tap tooltips with exact values. Make it sharp, modern and executive-ready. So the page is alive without input, on load the funnel pours in stage by stage and a tour cycles forever: it switches channels on the funnel, steps through the weeks on the trend chart, flags the anomaly on the heatmap, and moves the what-if sliders to show the projection, until the user touches a control. Save it as index.html in the current directory: one self-contained file with inline CSS and JavaScript only, no libraries or frameworks, and no external requests of any kind (no CDNs, web fonts, images or audio files). It must work when opened directly from disk, fit any window from a 390 px wide phone to a desktop, and run without console errors. 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
4 intermediate outputs, in the order the model wrote them.
No transcript was delivered with this result.
Cost breakdown
Headline cost $4.30, estimated from the token counts at list price.
| Tokens by type | Tokens | $ per 1M | Cost |
|---|---|---|---|
| Input (uncached) | 93 | $10.00 | $0.0009 |
| Cache write | 65,456 | $10.00 | $0.6546 |
| Cache read | 1,334,358 | $1.00 | $1.3344 |
| Output | 46,142 | $50.00 | $2.3071 |
| of which reasoning | 11,658 | billed as output |
- Total computed from tokens
- $4.2969
- Total reported by the harness
- $4.4606
- Difference
- OpenCode's own reported total ($4.4606) is shown beside the estimate; the list-price estimate is the headline, as on the other rows of this release.
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). The harness reports reasoning tokens separately from output; here output includes them (34,484 visible + 11,658 reasoning), and both bill at the output rate.
Prices: Project price table (vendor list prices), 2026-09-22.
Run notes
- frozen prompt (written): written 2026-10-05 for the MR category expansion (owner request: 500 finished runs in every category); category data_dashboards; 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: none
- Renderer: inline SVG in one self-contained HTML file (first <svg> at index.html line 29); no libraries, no CDN, no external assets, no canvas. No WebGPU or WebGL needed.
- Card video: the run's own 8 s session capture (video.mp4, H.264 yuv420p, 1280x720). Poster: the run's own render.png.
- Same frozen prompt as the Codex CLI + GPT-6 Astra row (dash_ecommerce_funnel__codex__gpt-6-astra__attempt1): two harnesses on one model, both unscored, so the pair is not a head-to-head result and no winner is claimed.




