$ replaybook

Model results

GPT-6 Luna

opencode-go/gpt-6-luna

Each row is an independent scenario cohort. Releases, providers and reasoning settings remain separate; these results are not pooled into a single model score.

Repairs, timing and cost

ScenarioReleaseProvider / reasoningRepairsMedian timeKnown costEvidence
028-nix-store-disk-pressure v1core 20261018.0.0OpenCode Go / Default 3/30 unavailable 124s Unknown Inspect attempts →
024-discourse-interrupted-deploy v1core 20261018.0.0OpenCode Go / Default 2/30 unavailable 88s Unknown Inspect attempts →
021-discourse-shared-uploads v1core 20261018.0.0OpenCode Go / Default 3/30 unavailable 140s Unknown Inspect attempts →
019-rust-fd-leak v1core 20261018.0.0OpenCode Go / Default 3/30 unavailable 109s Unknown Inspect attempts →
017-partial-rails-rollout v1core 20261018.0.0OpenCode Go / Default 3/30 unavailable 160s Unknown Inspect attempts →
015-sidekiq-poison-pill v1core 20261018.0.0OpenCode Go / Default 3/30 unavailable 146s Unknown Inspect attempts →
013-sidekiq-wrong-redis v3core 20261018.0.0OpenCode Go / Default 3/30 unavailable 68s Unknown Inspect attempts →
001-nginx-502-host v1core 20261018.0.0OpenCode Go / Default 3/30 unavailable 60s Unknown Inspect attempts →

Compare benchmark configurations → · All models