Original tests / 我们的原创实测
We wrote these prompts and ran each one once, identically, on Gemini 4 Hydrogen, Gemini 3.8 Flash, and Gemma 4 31B through the Gemini API free tier on 14 November 2026 (America/Denver). Hydrogen and 3.8 Flash used thinking_level=MEDIUM; Gemma 4 has no thinking level. We did not edit the models’ code. Playable frames show the raw HTML or SVG (Three.js CDN URLs were rewritten to a local copy so the iframe does not call the public internet). Dollar figures are equivalent paid-tier list prices — labelled “equivalent paid cost; ran on free tier.” A 429 or quota error stops that model for the rest of the batch; leftover tests wait until the free-tier quota resets at midnight Pacific. Free-tier prompts may be used by Google to improve products. This is not a benchmark: one run per prompt, our tasks, our settings.
Community same-prompt runs →
Free access guide

2 / 3 runs published
| Model | Status | Wall time | Input | Thinking | Output | Equiv. paid cost | Finish |
|---|
| Gemini 4 Hydrogen | broken | 2m 33s | 180 | 4,693 | 11,691 | $0.1231 * | MAX_TOKENS |
| Gemini 3.8 Flash | error | — | — | — | — | — | — |
| Gemma 4 31B | ran | 2m 10s | 152 | 1,538 | 2,033 | $0.0000 * | STOP |
* Equivalent paid cost; ran on free tier. Hydrogen $1.50 / $7.50 per 1M; 3.8 Flash $0.75 / $3.75 per 1M through 31 Dec 2026; Gemma listed free ($0.00). Output price includes thinking tokens.
Open playable results →

2 / 3 runs published
| Model | Status | Wall time | Input | Thinking | Output | Equiv. paid cost | Finish |
|---|
| Gemini 4 Hydrogen | ran | 54.8s | 159 | 1,409 | 9,841 | $0.0846 * | STOP |
| Gemini 3.8 Flash | error | — | — | — | — | — | — |
| Gemma 4 31B | ran | 1m 51s | 129 | 825 | 2,976 | $0.0000 * | STOP |
* Equivalent paid cost; ran on free tier. Hydrogen $1.50 / $7.50 per 1M; 3.8 Flash $0.75 / $3.75 per 1M through 31 Dec 2026; Gemma listed free ($0.00). Output price includes thinking tokens.
Open playable results →

2 / 3 runs published
| Model | Status | Wall time | Input | Thinking | Output | Equiv. paid cost | Finish |
|---|
| Gemini 4 Hydrogen | ran | 50.6s | 1,225 | 2,621 | 3,164 | $0.0452 * | STOP |
| Gemini 3.8 Flash | error | — | — | — | — | — | — |
| Gemma 4 31B | ran | 2m 1s | 358 | 914 | 2,407 | $0.0000 * | STOP |
* Equivalent paid cost; ran on free tier. Hydrogen $1.50 / $7.50 per 1M; 3.8 Flash $0.75 / $3.75 per 1M through 31 Dec 2026; Gemma listed free ($0.00). Output price includes thinking tokens.
Open playable results →

1 / 3 runs published
| Model | Status | Wall time | Input | Thinking | Output | Equiv. paid cost | Finish |
|---|
| Gemini 4 Hydrogen | ran | 1m 5s | 133 | 2,378 | 10,272 | $0.0951 * | STOP |
| Gemini 3.8 Flash | error | — | — | — | — | — | — |
| Gemma 4 31B | error | — | — | — | — | — | — |
* Equivalent paid cost; ran on free tier. Hydrogen $1.50 / $7.50 per 1M; 3.8 Flash $0.75 / $3.75 per 1M through 31 Dec 2026; Gemma listed free ($0.00). Output price includes thinking tokens.
Open playable results →

2 / 3 runs published
| Model | Status | Wall time | Input | Thinking | Output | Equiv. paid cost | Finish |
|---|
| Gemini 4 Hydrogen | ran | 57.8s | 146 | 1,543 | 11,112 | $0.0951 * | STOP |
| Gemini 3.8 Flash | error | — | — | — | — | — | — |
| Gemma 4 31B | ran | 2m 41s | 117 | 933 | 3,471 | $0.0000 * | STOP |
* Equivalent paid cost; ran on free tier. Hydrogen $1.50 / $7.50 per 1M; 3.8 Flash $0.75 / $3.75 per 1M through 31 Dec 2026; Gemma listed free ($0.00). Output price includes thinking tokens.
Open playable results →

2 / 3 runs published
| Model | Status | Wall time | Input | Thinking | Output | Equiv. paid cost | Finish |
|---|
| Gemini 4 Hydrogen | ran | 1m 14s | 162 | 3,216 | 15,470 | $0.1404 * | STOP |
| Gemini 3.8 Flash | error | — | — | — | — | — | — |
| Gemma 4 31B | ran | 2m 30s | 132 | 749 | 3,354 | $0.0000 * | STOP |
* Equivalent paid cost; ran on free tier. Hydrogen $1.50 / $7.50 per 1M; 3.8 Flash $0.75 / $3.75 per 1M through 31 Dec 2026; Gemma listed free ($0.00). Output price includes thinking tokens.
Open playable results →
What we did not run in this batch
A floor-plan-from-video test was on the candidate list. We did not run it: a walkthrough video would have used a large share of a small free-tier token budget.