Builder library

Pick a benchmark. Build your version.

Browse every released Builder test individually. Get its exact prompt and any released projects, Harness workflows and skills.

Browse all 12 tests Verify the benchmark evidence

12released benchmark tests, shown once each

Builder inventory

12 benchmark tests included

Exact prompts for every test, plus the released source projects, RemakeBench Harness and skills attached to each one.

Shader / WebGL

Infinite Cathedral

Prompt
4 projects

4 ready projects

Open test materials

Shader / WebGL

Neo-Gothic Storm City

Prompt
4 projects

4 ready projects

Open test materials

Spacecraft approaching a ringed planet in the Space Flight v1 collection

Three.js / browser

Space-Flight Game

Prompt
4 projects

3 ready + 1 reference-only

The GPT-5.6 Sol Ultra pre-request snapshot is source-only and labelled reference-only. Bundled ship-model copies retain CC BY 4.0 attribution and modification notices.

Open test materials

Campfire benchmark output in the Campfire Interactive Scene v1 collection

Three.js / browser

Campfire Under a Starry Night

Prompt
4 projects

3 ready + 1 reference-only

GPT-5.6 Sol xhigh is a source-only reference capture; inspect its per-output status before use.

Open test materials

Blender

Jungle Temple

Prompt
4 projects

4 ready projects

Open test materials

Blender

Shrine Village

Prompt
3 projects

3 ready projects

Open test materials

Blender

Oasis Outpost

Prompt
4 projects

4 ready projects

Open test materials

Blender

MacBook Cinematic

Prompt
3 projects

3 partial projects

Customer scenes are privacy-sanitized and unbranded; supplied reference stills, checkpoint files, branding, keyboard legends, and system-font-derived material are excluded.

Open test materials

Three.js / browser

JRPG Boss Battle

Prompt
3 projects

3 reference-only projects

Supplied GLB fixtures and target-video media are excluded. Each project includes asset-reconstruction instructions and is labelled reference-only.

Open test materials

Three.js / browser

Red Sands v1

Prompt
4 projects

4 ready projects

Release smoke proves clean installation, production build, and the fixture smoke route; it does not prove a human completed each authored quest.

Open test materials

Three.js / browser

Fighter Game v1

Prompt
4 projects

4 ready projects

Release smoke proves clean extraction, checksum verification, clean installation, production build, built-project start, and browser runtime boot; it is not a blind comparative evaluation or quality score.

Open test materials

Unity

Shrine Exploration

Prompt
Unity project
Harness
4 skills

Completed Unity source project

Open test materials

Public benchmark evidenceVerify what happened.

Results, costs, tokens, workflow time, run ledgers, methodology, evidence, and known gaps remain free to inspect.

Free public comparisons

Latest published comparisons

Inspect the published outcomes first. Evidence details and known gaps stay accessible below each comparison.

Samurai looking up through the illuminated torii in Shrine Exploration

Launch 004 · public comparison

32 recorded runsPublished 2026-07-31

Claude Opus 5 across nine game-building tests

Off-road Mud Game · Space Flight Stage 2 — Landfall · Neo-Gothic Flooded City · Infinite Cathedral · MacBook Pro · JRPG Boss Battle · Jungle Temple · Oasis Outpost · Shrine Exploration

Evidence available & known gaps

Evidence available

Cost, workflow time, token, stack, artifact, validation, and caveat ledgers
Nine frozen tests with 32 exact featured records
Fourteen hash-bound Launch 004 presentation recordings
Public Shrine receipts, disclosed prompt privacy findings, platform limitations, and exclusion proof
Published owner benchmark assessment

Still missing

Companion video episode metadata (deferred after benchmark publication)
Repeat-run reliability evidence
Complete Shrine subagent token mix and exact total cost
WebGL gameplay, physical-input, and audible-output verification
Windows runtime, gameplay, physical-input, and audio verification

Silver laptop illuminated against a dark studio background

Launch 003 · public comparison

42 recorded runsPublished 2026-07-23

I Gave Qwen3.8 Max, Kimi K3, GPT-5.6 and Fable 5 the Same 9 Game Tests

Neo-Gothic Storm City · MacBook-Class Cinematic · JRPG Boss Battle · Campfire Under a Starry Night · Space Flight Game · Shrine Village · Infinite Cathedral · Oasis Outpost · Jungle Temple

Evidence available & known gaps

Evidence available

Cost, workflow time, token, stack, artifact, and caveat ledgers
Nine frozen Qwen tests with a 42-record comparison pool
Nine public Qwen media recordings and linked artifacts
Published owner editorial assessment and official video

Still missing

Minimum community preference sample
Repeat-run reliability evidence
A completed Qwen MacBook agent run

Knight facing an armored boss in a volcanic arena

Launch 002 · public comparison

33 recorded runsPublished 2026-07-19

Kimi K3 vs GPT-5.6 SOL vs Claude Fable 5: 9 Game Tests

Neo-Gothic Storm City · MacBook-Class Cinematic · JRPG Boss Battle · Campfire Under a Starry Night · Space Flight Game · Shrine Village · Infinite Cathedral · Oasis Outpost · Jungle Temple

Evidence available & known gaps

Evidence available

Cost, workflow time, token, stack, and artifact ledgers
Nine frozen tests with 33 featured run records
Seven anonymous Kimi-versus-Fable blind matchups
Published editorial assessment and official video

Still missing

Minimum community preference sample
Repeat-run reliability evidence
One reconciled Kimi MacBook API request

Spacecraft approaching a ringed planet through an asteroid field

Launch 001 · public comparison

28 recorded runsPublished 2026-07-15

GPT-5.6 vs Claude Fable 5: Same Prompts, 7 AI Game-Dev Tests

Infinite Cathedral Corridor · Neo-Gothic Storm City · Space Flight Game · Campfire Under a Starry Night · Jungle Temple · Shrine Village · Oasis Outpost

Evidence available & known gaps

Evidence available

Cost, workflow time, token, stack, and artifact ledgers
Seven standardized Fable-versus-Sol comparisons
Anonymous blind exploration; sign-in only to record a vote
Published editorial assessment and official video

Still missing

Minimum community preference sample
Repeat-run reliability evidence

11 scoped collections

Versioned benchmark collections

Stable scopes over every measured run. Open one to see which models were tested, how each attempt went, and where the public evidence still has gaps.

Harness v1 · Builder projects v1

4 measured runsPublished 2026-08-08

Red Sands v1

Evidence available & known gaps

Evidence available

Four selected measured run ledgers
Exact frozen prompt identity
Per-run quest source and generated voice artifact identities
Four reconstructed projects with clean install, production build, and browser smoke receipts

Still missing

Human start-to-finish playthrough for every run
Blind evaluation and quality scoring
Repeat-run reliability evidence

Harness v1 · Builder projects v1

4 measured runsPublished 2026-08-08

Fighter Game v1

Evidence available & known gaps

Evidence available

Four selected measured run ledgers
Exact frozen prompt identity
Per-run arena-geometry artifact identities
Exact clean GPT-5.6 Sol rerun with superseded contaminated attempt excluded

Still missing

Blind evaluation and scored side-by-side review
Standardized FPS and hitch receipts
Repeat-run reliability evidence

Harness v1

8 measured runsPublished 2026-07-19

Space Flight v1

Explorable space-flight game

Evidence available & known gaps

Evidence available

Eight measured run ledgers
Public browser builds
Cost, token, and workflow timing

Still missing

Browser/hardware environment
Local FPS
Final capture metadata
Minimum blind-vote sample

Astronaut approaching a landed spacecraft in the Space Flight v2 collection

Harness v2

4 measured runsPublished 2026-07-31

Space Flight v2

Space Flight Stage 2 — Landfall

Evidence available & known gaps

Evidence available

Four measured run ledgers
Four public presentation recordings
Cost, token, workflow timing, and artifact receipts

Still missing

Repeat-run reliability evidence
Standardized gameplay and physical-input verification
Comparable local FPS measurements

Harness v1

19 measured runsPublished 2026-07-19

Shader Frontier v1

Infinite cathedral corridor shader · Neo-gothic storm city shader

Evidence available & known gaps

Evidence available

Nineteen measured run ledgers
Shader artifacts with SHA-256
Cost, token, and workflow timing

Still missing

Initial and optimized local FPS
RTX Pro 6000 final-capture metadata
Minimum blind-vote sample

Harness v1 · Capture protocol v1

24 measured runsPublished 2026-07-19

Blender Voxel Worlds v1

High-voxel-density Jungle Temple diorama · High-voxel-density Oasis Outpost diorama · High-voxel-density Shrine Village diorama

Evidence available & known gaps

Evidence available

Twenty-four measured run ledgers
Primary artifact SHA-256 records
Standardized turntable media

Still missing

Recorded Blender/render environment for original runs
Repeat-run evidence
Minimum blind-vote sample

Harness v1

8 measured runsPublished 2026-07-19

Campfire Interactive Scene v1

Campfire under the stars

Evidence available & known gaps

Evidence available

Seven measured run ledgers
Public browser builds
Cost, token, and workflow timing

Still missing

Browser/hardware environment
Local FPS for several runs
Final capture metadata
Minimum blind-vote sample

Harness v1

5 measured runsPublished 2026-07-19

MacBook Cinematic v1

MacBook-class cinematic ad scene

Evidence available & known gaps

Evidence available

Five measured run ledgers
Public recorded outputs
Cost, token, and workflow timing

Still missing

One reconciled Kimi API request
Repeat-run reliability evidence
Minimum blind-vote sample

Harness v1

5 measured runsPublished 2026-07-19

JRPG Boss Battle v1

JRPG boss battle

Evidence available & known gaps

Evidence available

Five measured run ledgers
Public recorded outputs
Cost, token, and workflow timing

Still missing

Repeat-run reliability evidence
Minimum blind-vote sample

Harness v1

8 measured runsPublished 2026-07-19

Interactive Voxel Viewer v1

Interactive Three.js voxel map viewer

Evidence available & known gaps

Evidence available

Seven measured run ledgers
Primary artifact integrity records
Cost, token, and workflow timing

Still missing

Standardized public capture set
Repeat-run reliability evidence
Minimum blind-vote sample

Result v1 · public evidence

1 measured runsPublished 2026-07-31

Shrine Exploration v1

Shrine Exploration

Evidence available & known gaps

Evidence available

One measured long-horizon result record
12 → 14 → 14 / 40 visual score history
$204.17+ parent-session API list-price lower-bound receipt
Three hosted public-player archives with streamed read-back and a WebGL startup route
Target-specific Builder-source artifact identities

Still missing

Complete subagent token mix and exact total cost
Repeat-run reliability evidence

Community access is free

Discuss the result. Build the next one.

Join after signing in. Builder adds released reproduction components and private Builder Lab; it is not required for the community Discord or public benchmark evidence.

Join the verified member Discord →