Measured harness ledgerPublic result
Claude Fable 5.1MacBook-class cinematic ad scene — Claude Fable 5.1 Max
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.
Max reasoningHeadline result
- Workflow cost
- $171.34
- Wall-clock
- 5h 19m 27.3s wall-clock
- Processed tokens
- 109.82M processed
- Record state
- partial_active_session_snapshot_with_supplied_validation
Public summary
Claude Fable 5.1 Max partial_active_session_snapshot_with_supplied_validation ledger: 5h 19m 27.3s wall-clock, 109.82M processed, and $171.34 Official Anthropic API-list-price equivalent from the supplied active-session snapshot, not a Claude Code subscription invoice..
Run identity and stack
- Result ID: macbook-cinematic-blender-fable-5.1-max
- Technical model: claude-fable-5-1
- Provider: Anthropic Claude Code
- Stack: Anthropic Claude Code
- Stack: Technical model/configuration: claude-fable-5-1
- Stack: Blender MCP
- Stack: Cinematic product scene
- Stack: Harness v1 MacBook validator
- Stack: Requested tool profile: blender-mcp
Cost basis
- The supplied receipt reports live verification of the listed rates.
- All-5-minute cache-write alternative: $136.59.
Primary artifact integrity
- Kind: blender-cinematic-product-ad-scene
- Path: artifacts/macbook-cinematic-blender-fable-5.1-max/macbook_cinematic_final.blend
- SHA-256: 2282a6e5ce6a849801dc0b9a6125d20b6a75c5c774ca032c694cedcf958cd67b
Validation evidence
- Result: PASS, 49/49 supplied checks
- Path: artifacts/macbook-cinematic-blender-fable-5.1-max/evidence/validation_log.json
- SHA-256: 79dbff1d3b31ade38247bbc1fab7d48da16eae88ff3ed4fe317b5839e7271e02
- Validator SHA-256: 54591de8d4ef2953c6f7e2fe3e4f4b3d9407b953049cc858b1012afaf9424d51
Recorded caveats
- The metrics are an active-session snapshot rather than final accounting.
- Wall-clock includes Blender-side work, tool execution, render waits, and other workflow overhead; it is not model-only compute time.
- Output tokens include hidden reasoning, generated code, and tool-call JSON, not only visible prose.
- The supplied 49/49 validation log was JSON-validated and fixture-checked, but Blender was unavailable for an independent reopen, validator rerun, or render.
- The six stills are supplied model-run rubric evidence, not blind-evaluation evidence.
- No render-hardware identity or blind-evaluation record is supplied.
Visible evidence gaps
- final inactive-session token and cost receipt
- independent Blender reopen and headless validator rerun
- render-hardware receipt
- independent final capture
- 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.