Measured harness ledgerPublic result
GLM 5.3 FlashInfinite cathedral corridor shader — GLM 5.3 Flash Max
Generate an infinite stained-glass cathedral corridor shader with convincing depth, architectural repetition, light, and motion.
Max reasoningHeadline result
- Workflow cost
- $0.59
- Wall-clock
- 1h 26m 0.031s wall-clock
- Processed tokens
- 8.32M processed
- Record state
- complete
Public summary
GLM 5.3 Flash Max complete ledger: 1h 26m 0.031s wall-clock, 8.32M processed, and $0.59 API-equivalent estimate at current Z.ai list pricing; not a historical invoice.
Run identity and stack
- Result ID: infinite-cathedral-stained-glass-shadertoy-glm-5.3-flash-max-2026-08-21
- Technical model: glm-5.3-flash
- Provider: Provider not separately recorded
- Client: OpenCode CLI 1.18.18
- Stack: OpenCode CLI 1.18.18
- Stack: Technical model/configuration: glm-5.3-flash
- Stack: ShaderToy fragment shader
- Stack: Harness v1 infinite cathedral prompt
Cost basis
- Current OpenRouter discounted equivalent: $0.29.
Primary artifact integrity
- Kind: manifest-verified-artifact
- Path: artifacts/infinite-cathedral-stained-glass-shadertoy-glm-5.3-flash-max-2026-08-21/source/main.glsl
- SHA-256: bfed42e8ca7cc5380966798ab3d3b32cd1bea7fbe6d58594ff725fb334a7dc7d
Validation evidence
- Result: complete
- Path: artifacts/infinite-cathedral-stained-glass-shadertoy-glm-5.3-flash-max-2026-08-21/verification/summary.public.json
- SHA-256: 7e5fa8071e21a4d0fe8a67c0a343fce736ecbd1284079d41866cdacb092b52ae
Recorded caveats
- The run required repeated user-message continuations because the OpenCode endpoint frequently exited or stalled before task completion.
- OpenCode reported $0.00 for the free preview endpoint; recovered session usage is priced separately as a current API-equivalent estimate, not a historical charge.
- No blind evaluation or quality score is attached.
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.
