Measured harness ledgerPublic result
GPT-6 Astra planner/author + Jev 1.13 actorApple Stem — GPT-6 Astra planner/author + Jev 1.13 actor Unreported
Pinch an apple by the stem and lift it with a Wuji hand on an FR3 arm in SuperDex, without contacting the apple body.
Unreported reasoningHeadline result
- Workflow cost
- $43.54
- Wall-clock
- Not recorded
- Processed tokens
- Not recorded
- Record state
- disclosed_negative_result_capped_no_pass
Public summary
GPT-6 Astra planner/author + Jev 1.13 actor Unreported disclosed_negative_result_capped_no_pass ledger: Not recorded wall-clock, Not recorded, and $43.54 Recorded provider cost.
Run identity and stack
- Result ID: dex-apple-stem-superdex-astra-jev-learn-2026-09-21
- Technical model: GPT-6 Astra planner/author + Jev 1.13 actor
- Provider: Provider not separately recorded
- Stack: Technical model/configuration: GPT-6 Astra planner/author + Jev 1.13 actor
- Stack: SuperDex fingertip fixture
- Stack: Harness apple-stem prompt
Validation evidence
- Path: artifacts/dex-apple-stem-superdex-astra-jev-learn-2026-09-21/validation/judge-replay.public.json
- SHA-256: 600b8d93a9ec66620f021a38946b21a204bb0f4070d6a496ebf2f204b8966bfe
Recorded caveats
- NOT COMPARABLE WITH THE CODING-AGENT COHORTS OR WITH BARE DECISION-LOOP ROWS. This is a jev-astra-hybrid run: GPT-6 Astra and Jev 1.13 were coupled under a protocol that is not the one-shot locked-prompt coding-agent track and not the two-call-per-cycle decision-loop track. cohort_eligible is false. Harry's criterion (same verdict as Astra-alone AND cheaper on API-equivalent) is recorded in the track README; it is not a league-table standing.
- DISCLOSED NEGATIVE RESULT. Capped 12/12, no PASS. Do not resume.
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.