Explorable space-flight game — Qwen3.8 Max Preview Unreported

Measured harness ledgerPublic result

Explorable space-flight game — Qwen3.8 Max Preview Unreported

Build a responsive browser space-flight game with flight controls, a coherent star-system environment, lighting, assets, and a playable game loop.

Unreported reasoningHeadline result

Workflow cost: $6.95
Wall-clock: 38m 43.2s wall-clock
Processed tokens: 4.44M processed
Record state: partial_token_timing_runtime_artifact_ledger_orientation_failure

Public summary

Qwen3.8 Max Preview Unreported partial_token_timing_runtime_artifact_ledger_orientation_failure ledger: 38m 43.2s wall-clock, 4.44M processed, and $6.95 Third-party NanoGPT API-list-price equivalent for the measured token mix; not an Alibaba Token Plan cash charge.

Run identity and stack

Result ID: space-flight-game-qwen3.8-max-preview
Technical model: qwen3.8-max-preview
Provider: Alibaba Cloud Model Studio
Client: Qwen Code 0.20.0
Stack: Alibaba Cloud Model Studio
Stack: Qwen Code 0.20.0
Stack: Technical model/configuration: qwen3.8-max-preview
Stack: Three.js / Vite game
Stack: Generated spaceship assets
Stack: Harness v1 space-flight prompt
Stack: Requested tool profile: base shell/file tools

Cost basis

NanoGPT published Qwen3.8 Max Preview rates of $1.50/M input and $5.00/M output. It did not publish a separate cache-read discount, so all measured input tokens use the same input rate. The actual run used Alibaba Token Plan Personal Lite; $6.95 is a third-party API-list-price equivalent, not a first-party Alibaba price, an itemized provider charge, or provider cost.
Qualified API-list-price equivalent.

Primary artifact integrity

Kind: threejs-space-flight-game-source-entry-point
Path: artifacts/space-flight-game-qwen3.8-max-preview/source/src/main.js
SHA-256: 1307a0deeda5ea1b9e2cf39977dbea4bc1262eb440f2618ecf1d4dc97cd5cea0

Recorded caveats

Wall-clock is end-to-end workflow latency, not model-only compute, and includes tool execution and waits within the recorded prompt window.
Output tokens include Qwen-accounted thoughts/reasoning, visible prose/code, and tool-related output; 40,727 thought tokens are a subset of the 82,279 output tokens.
The 120 FPS figure is a post-run HUD-reported measurement on an Apple M1 Pro / ANGLE Metal browser session, not a standardized GPU benchmark or model-supplied performance capture.
The $6.95 estimate uses NanoGPT's published third-party Qwen3.8 Max Preview API rates, not an Alibaba first-party pay-as-you-go price or a task-level subscription charge.
The task-isolated Credit estimate is provisional and after-only; its weekly meter cannot be allocated to this task alone.
The result has no quality score because the ship orientation is confirmed wrong and the model did not produce its own final capture.
No blind-evaluation record or standardized RTX render was supplied.

Visible evidence gaps

a fresh infrastructure-corrected retry with a model-accessible hardware-backed capture path
correct fighter orientation and improved gameplay framing
standardized local FPS measurement receipt
standardized RTX render
blind-evaluation record

Public result only

This result keeps its public summary and evidence, but it does not currently have a matching Builder test with prompts, projects, Harness workflows or skills.

Browse Builder Return to Qwen3.8 Max Preview