Measured harness ledgerPublic result
Qwen3.8 MaxMacBook-class cinematic ad scene — Qwen3.8 Max xhigh
Create one polished MacBook-class product-ad shot in Blender with a modeled device, legible industrial detail, intentional materials, lighting, camera movement, and a validator-ready scene.
xhigh reasoningHeadline result
- Workflow cost
- $24.99
- Wall-clock
- 2h 51m 49.9s wall-clock
- Processed tokens
- 82.89M processed
- Record state
- artifact_validator_pass_metrics_partial_quota_interrupted
Public summary
Qwen3.8 Max xhigh artifact_validator_pass_metrics_partial_quota_interrupted ledger: 2h 51m 49.9s wall-clock, 82.89M processed, and $24.99 API-equivalent accounting only, not an itemized Token Plan cash charge.
Run identity and stack
- Result ID: macbook-cinematic-blender-qwen3.8-max-xhigh
- Technical model: Qwen 3.8 Max
- Provider: Alibaba Cloud Model Studio Token Plan via Qwen Code
- Client: Qwen Code 0.21.7 installed / 0.21.6 transcript-recorded
- Stack: Alibaba Cloud Model Studio Token Plan via Qwen Code
- Stack: Qwen Code 0.21.7 installed / 0.21.6 transcript-recorded
- Stack: Technical model/configuration: Qwen 3.8 Max
- Stack: Blender MCP
- Stack: Cinematic product scene
- Stack: Harness v1 MacBook validator
Cost basis
- Qwen Code records no itemized per-run cash cost under this Token Plan.
Primary artifact integrity
- Kind: blender-cinematic-scene
- Path: artifacts/macbook-cinematic-blender-qwen3.8-max-xhigh/macbook_cinematic_final.blend
- SHA-256: f1d7a5e482656e3d7440638326bb74aee83457b8a358c5921cb1ea1bf809af43
Validation evidence
- Result: 49/49 checks recorded
- Path: artifacts/macbook-cinematic-blender-qwen3.8-max-xhigh/validation_log.json
- SHA-256: c6636a210a52408248a0157f71833ce80adbef7760dd98a4c56b2531ac0222c3
Recorded caveats
- The final blend and 49/49 supplied validator log are present, but the run cannot be represented as a fully canonical metrics window because quota exhaustion forced a synthetic EOF measurement boundary.
- The objective validator was not independently rerun in this archival pass.
- Wall-clock is end-to-end latency, not model-only compute; it includes tools, installs, render work, browser checks, and idle gaps.
- One benchmark user turn generated 237 API requests, including five same-model visual-inspector agents.
- No cache-write counter, quota delta, blind-evaluation record, or standardized runtime-performance measurement is available.
- Visual quality is deliberately not scored in this ledger.
Visible evidence gaps
- natural post-run metrics boundary
- independent validator rerun with Blender version and render 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.