From 79f4cbac8ecf3694856737671df8a7142c42d8dc Mon Sep 17 00:00:00 2001 From: Pedro Rodrigues Date: Tue, 28 Jul 2026 16:20:08 +0100 Subject: [PATCH] feat: bump Opus experiments to Claude Opus 5 Swaps the two Opus 4.8 experiments to Claude Opus 5 (claude-opus-5) and updates the matching README reference. Pinned Claude Code CLI (2.1.191) passes the explicit model ID through unchanged; sanity run passed 7/7. Closes AI-968 Co-Authored-By: Claude Fable 5 --- README.md | 2 +- ...de-opus-4.8-no-skills.ts => claude-code-opus-5-no-skills.ts} | 2 +- experiments/{claude-code-opus-4.8.ts => claude-code-opus-5.ts} | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) rename experiments/{claude-code-opus-4.8-no-skills.ts => claude-code-opus-5-no-skills.ts} (93%) rename experiments/{claude-code-opus-4.8.ts => claude-code-opus-5.ts} (94%) diff --git a/README.md b/README.md index 7662b01a..31527982 100644 --- a/README.md +++ b/README.md @@ -51,7 +51,7 @@ Run selected evals across multiple experiments: ```bash pnpm eval -- \ --experiment claude-code-sonnet-5 \ - --experiment claude-code-opus-4.8 \ + --experiment claude-code-opus-5 \ --eval resolve-dataapi-001-empty-results \ --eval investigate-auth-001-deleted-user-access ``` diff --git a/experiments/claude-code-opus-4.8-no-skills.ts b/experiments/claude-code-opus-5-no-skills.ts similarity index 93% rename from experiments/claude-code-opus-4.8-no-skills.ts rename to experiments/claude-code-opus-5-no-skills.ts index 09f115fb..a93df1d8 100644 --- a/experiments/claude-code-opus-4.8-no-skills.ts +++ b/experiments/claude-code-opus-5-no-skills.ts @@ -9,7 +9,7 @@ import { localStackRuntime } from '@supabase-evals/sandbox'; export default defineExperiment({ suite: ['no-skills'], agent: claudeCodeAgent({ - model: 'claude-opus-4-8', + model: 'claude-opus-5', reasoningEffort: 'high', }), runtime: platformLiteRuntime({ diff --git a/experiments/claude-code-opus-4.8.ts b/experiments/claude-code-opus-5.ts similarity index 94% rename from experiments/claude-code-opus-4.8.ts rename to experiments/claude-code-opus-5.ts index e27a3d6d..0ffe0823 100644 --- a/experiments/claude-code-opus-4.8.ts +++ b/experiments/claude-code-opus-5.ts @@ -9,7 +9,7 @@ import { localStackRuntime } from '@supabase-evals/sandbox'; export default defineExperiment({ suite: ['benchmark'], agent: claudeCodeAgent({ - model: 'claude-opus-4-8', + model: 'claude-opus-5', reasoningEffort: 'high', }), runtime: platformLiteRuntime({