Measured harness ledgerPublic result
Claude Opus 5

High-voxel-density Shrine Village diorama — Claude Opus 5 Max

Create a complete inspectable Blender voxel diorama of a shrine village at a fixed 0.05-unit voxel resolution through the Blender MCP workflow.

Max reasoningHeadline result
Workflow cost
$29.67
Wall-clock
41m 47.5s wall-clock
Processed tokens
30.88M processed
Record state
partial_token_timing_artifact_ledger
Public summary

Claude Opus 5 Max partial_token_timing_artifact_ledger ledger: 41m 47.5s wall-clock, 30.88M processed, and $29.67 First-party Claude API list-price equivalent, not a subscription invoice.

Run identity and stack
  • Result ID: shrine-village-opus-5-max
  • Technical model: claude-opus-5
  • Provider: Anthropic Claude Code with Blender MCP
  • Stack: Anthropic Claude Code with Blender MCP
  • Stack: Technical model/configuration: claude-opus-5
  • Stack: Blender MCP
  • Stack: 0.05-unit voxel grid
  • Stack: Inspectable .blend artifact
Cost basis
  • Primary cost uses Exact source-reported split: 100% one-hour cache writes..
  • All-5-minute cache-write alternative: $27.93.
Primary artifact integrity
  • Kind: blender-high-voxel-density-diorama
  • Path: artifacts/shrine-village-opus-5-max/shrine_village.blend
  • SHA-256: ed36cfff739ec9ec6586785e74ecea75d34646ce7166d69c82e96fc3fad2a3d4
Recorded caveats
  • Wall-clock is end-to-end latency, not model-only compute; the supplied receipt includes tool execution, Blender MCP round-trips, rendering, and user idle time.
  • Output tokens include hidden reasoning, code, and tool-call JSON, not only visible prose.
  • Cache reads are 97.2% of processed-token volume but bill at 0.1× input, so processed tokens substantially overstate cost.
  • The source receipt is a snapshot and can grow if rerun later in the still-active session.
  • The preview render was visually inspected during archival as a coherent scene, but no visual-quality score or blind comparison has been assigned.
  • The artifact was not independently reopened or validator-tested during archival because Blender is unavailable in this environment.
Visible evidence gaps
  • independent Blender load or validator receipt
  • standardized final-render hardware and render-settings 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.

RemakeBenchResearch console