Public evidence snapshot

rr_local_models_benchmark_fixture

This build-owned snapshot keeps review provenance available to anonymous learners even when the original editorial repository or pull request is private or renamed.

Course

openclaw-local-models-performance-cost

Lesson

openclaw-local-models-performance-cost.benchmark

Reviewer

academy-specification-review

Method

source-review

Outcome

approved

Reviewed

2026-07-29

Content commit

694734c7a8002480a8a926d9821482c5b34a73ce

Dependency snapshot

0d5124f0d2ef5eb3ffd5fc55b5e068c412f1abd370ccb02d6708b37f7b2a2d57

Recorded sources

1. openclaw-docs-multi-agent — openclaw-cc570a74532-models
   Snapshot SHA-256: 98cd13cde14ee640411b00f2399cb589718526c3cb8b16dbe64b5e07d58b4ed0
   Recorded canonical reference: https://github.com/openclaw/openclaw/blob/cc570a74532e4907e5f1f8090dd05fc7a1ddfc27/docs/concepts/models.md
2. openclaw-docs-multi-agent — openclaw-cc570a74532-model-failover
   Snapshot SHA-256: 06d31b01773e4d082054136bd2071d5589bf6c522a2ccf76904d02d51c5aabfa
   Recorded canonical reference: https://github.com/openclaw/openclaw/blob/cc570a74532e4907e5f1f8090dd05fc7a1ddfc27/docs/concepts/model-failover.md

Limitations

• Approval covers the original workload, benchmark, routing, quota, and fallback method for 2026.7.1; it makes no universal provider, model, GPU, performance, privacy, or savings claim.

Original review reference: https://github.com/openclaw-academy/learn/pull/20