Measured harness ledgerPublic result
Kimi K3F-117 Stealth Jet — Kimi K3 Max
Build an interactive Three.js F-117 stealth-jet experience under the pinned Harness task and fixture contract.
Max reasoningHeadline result
- Workflow cost
- $6.34
- Wall-clock
- 1h 24m 21.519s wall-clock
- Processed tokens
- 11.04M processed
- Record state
- partial_reconciled_metrics_and_static_source_ledger
Public summary
Kimi K3 Max partial_reconciled_metrics_and_static_source_ledger ledger: 1h 24m 21.519s wall-clock, 11.04M processed, and $6.34 Token-only API-equivalent estimate, not the marginal Kimi Code subscription charge.
Run identity and stack
- Result ID: f117-stealth-jet-threejs-kimi-k3-max
- Technical model: kimi-code/k3
- Provider: Kimi Code
- Stack: Kimi Code
- Stack: Technical model/configuration: kimi-code/k3
- Stack: Three.js flight experience
- Stack: Harness v1 F-117 prompt
Cost basis
- The run used an Allegretto subscription ($39/month); no marginal per-run cash charge is recorded.
Primary artifact integrity
- Kind: interactive-threejs-f117-model
- Path: artifacts/f117-stealth-jet-threejs-kimi-k3-max/source/main.js
- SHA-256: 2aa84ee9322fedab394ff897b09dd9a9b01fbb9ff19bb5a01327f9933774f00c
Validation evidence
- Path: artifacts/f117-stealth-jet-threejs-kimi-k3-max/validation/static-source-check.public.json
- SHA-256: d04383cb8b15aeec84a46ef63f49b9bb4c9651fb73b718b53ec696e9b1e9ece4
Recorded caveats
- The receipt does not record the user-turn count, so this ledger does not claim a confirmed clean one-shot protocol.
- Metrics are partial because the final in-window compaction request has a usage record without a completed-step record; its tokens are included once and exact reconciliation is documented.
- Wall-clock is end-to-end latency, including tools, installs, browser checks, and waits; it is not model-only compute time.
- Output includes code and tool-call payloads; a separate reasoning-token count was not available.
- Cache-read tokens are discounted, so processed-token volume overstates effective cost.
- The $6.34 figure is an API-equivalent estimate based on rates verified by the source receipt, not an actual marginal subscription charge.
- No independent browser acceptance run, local FPS measurement, hardware/viewport receipt, quota capture, or blind evaluation was supplied.
Visible evidence gaps
- explicit user-turn protocol receipt or clean one-user-turn rerun
- independent browser acceptance run
- local FPS, browser, viewport, and hardware receipt
- 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.