Add GP earning benchmark (5-loop iterative)#37
Open
classicrob wants to merge 1 commit intoMaxBittker:mainfrom
Open
Add GP earning benchmark (5-loop iterative)#37classicrob wants to merge 1 commit intoMaxBittker:mainfrom
classicrob wants to merge 1 commit intoMaxBittker:mainfrom
Conversation
New Harbor benchmark task where agents earn as much gold as possible. 5 loops with fresh sub-agents per loop, learnings.md as the handoff document between loops. Each loop gets 5 bots (level 50 all skills) and 10,000 game ticks per script. No pickpocketing. Also adds gemini-flash, gpt-5.3-codex, and kimi models to run.sh. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
gp-10k-ticks: agents earn as much gold as possible across 5 iterative loopslearnings.mdandgp_results.jsoncarry forwardgemini-flash(gemini-3-flash-preview),codex53(gpt-5.3-codex), andkimimodels to run.shNew files
benchmark/shared/gp_loop_instruction.md— per-loop instruction for sub-agentsbenchmark/shared/generate_gp_saves.ts— generates 25 save files (5 bots × 5 loops)benchmark/shared/check_gp.ts— verifier that reads per-loop GP results + verifies inventoryTest plan
bun benchmark/generate-tasks.tsgeneratesgp-10k-ticks/with correct instruction, Dockerfile, and verifierbenchmark/run.sh -t gp-10k-ticks -m gemini-flash -m codex53🤖 Generated with Claude Code