Measured harness ledgerPublic result
DeepSeek V4 ProMacBook-class cinematic ad scene — DeepSeek V4 Pro 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
- $0.84
- Wall-clock
- 9244.7s wall-clock
- Processed tokens
- 84.38M processed
- Record state
- complete_token_timing_artifact_render_and_validation_ledger
Public summary
DeepSeek V4 Pro Max complete_token_timing_artifact_render_and_validation_ledger ledger: 9244.7s wall-clock, 84.38M processed, and $0.84 API-equivalent accounting.
Run identity and stack
- Result ID: macbook-cinematic-deepseek-v4-pro-max
- Technical model: DeepSeek V4 Pro
- Provider: OpenCode Go
- Client: OpenCode
- Stack: OpenCode Go
- Stack: OpenCode
- Stack: Technical model/configuration: DeepSeek V4 Pro
- Stack: Blender MCP
- Stack: Cinematic product scene
- Stack: Harness v1 MacBook validator
Primary artifact integrity
- Kind: blender-scene
- Path: artifacts/macbook-cinematic-deepseek-v4-pro-max/macbook_cinematic_final.blend
- SHA-256: 9dee43e414cb7851f4e24e4964f0c438e3d579c3b7bc07299f96072466bc408a
Validation evidence
- Result: PASS
Recorded caveats
- Wall-clock includes Blender MCP/tool execution and is not model-only compute.
- Two benchmark continue recoveries make this a disclosed three-turn protocol deviation.
- Seven additional exact continue recoveries were required during the post-generation metrics audit and are outside the benchmark window.
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.