Measured harness ledgerPublic result
DeepSeek V4 FlashFighter Game v1 — DeepSeek V4 Flash Max
Improve the complete arena environment of a fixed Three.js fighting game through Blender MCP without changing the fighters, combat, HUD, effects, or audio.
Max reasoningHeadline result
- Workflow cost
- Not recorded
- Wall-clock
- 2h 06m 19.9s wall-clock
- Processed tokens
- 65.65M processed
- Record state
- partial_multi_user_turn_reconciled_metrics_artifact_ledger
Public summary
DeepSeek V4 Flash Max partial_multi_user_turn_reconciled_metrics_artifact_ledger ledger: 2h 06m 19.9s wall-clock, 65.65M processed, and Not recorded.
Run identity and stack
- Result ID: courtyard-arena-blender-mcp-deepseek-v4-flash-max
- Technical model: deepseek-v4-flash
- Provider: opencode-go
- Stack: opencode-go
- Stack: Technical model/configuration: deepseek-v4-flash
- Stack: Blender MCP
- Stack: Three.js fighting-game fixture
- Stack: Harness v1 frozen prompt
Cost basis
- No measured or estimated run cost is recorded.
Primary artifact integrity
- Kind: manifest-verified-artifact
- Path: artifacts/courtyard-arena-blender-mcp-deepseek-v4-flash-max/source/web/models/arena.glb
- SHA-256: 11e933e1f0561c143c3ee90b3a904160a5c430d9c170c9b2d1f6c8d802595faa
Recorded caveats
- Two user turns are recorded where the fixture protocol is one fixed prompt.
- The candidate input boundary is unverified; the run directory contained withheld fixture material, so no closed-book or blind-vote-eligibility claim is made.
- Wall-clock is end-to-end workflow latency, not model-only compute.
- Generated output includes both visible and reasoning output in this OpenCode receipt.
- Cache reads are discounted, so total processed tokens overstate cost.
- The recorded subtotal is not an actual subscription invoice; tool services, quota, and marginal cash charge are not available.
- No independent Blender reopen, runtime replay, FPS/hitch measurement, console receipt, or blind evaluation is recorded.
Visible evidence gaps
- verified closed-book candidate input boundary
- independent runtime replay and console receipt
- standardized FPS and hitch 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.