{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.16613", "kind": "paper", "url": "https://arxiv.org/abs/2606.16613"}, {"id": "gh:sakanaai/coffeebench", "kind": "repo", "url": "https://github.com/SakanaAI/CoffeeBench"}], "authors": ["Issa Sugiura", "Daichi Hattori", "Kazuo Araragi", "Keita Ogawa", "Shota Onose", "Taro Makino", "Teppei Usuki", "Takashi Ishida"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.16613", "source_url": "https://arxiv.org/abs/2606.16613"}, "categories": ["Agents", "Language & Knowledge", "Logistics & Operations"], "description": {"en": "CoffeeBench evaluates LLM agents as a coffee roaster in a 90-day multi-agent economy with fixed reference agents, measuring cumulative net income through autonomous communication, negotiation, and transactions."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.16613", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.16613", "title": "CoffeeBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.16613", "modality": null, "name": "CoffeeBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_coffeebench_22526372", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "49c3ba4aec201f583b74e0c9a78f5d77ad829043b8abdcabd596b6b2d34f933d", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.16613"}, "publisher": null, "released": "2026-06-15", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.16613", "source_url": "https://arxiv.org/abs/2606.16613"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-16613", "source": "claire_radar", "source_benchmark_id": "2606.16613", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Logistics & Operations"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/SakanaAI/CoffeeBench", "githubScope": "benchmark_repo", "githubStars": 37, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 11, "hfPaperUrl": "https://huggingface.co/papers/2606.16613", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-01", "state": "stale"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "b4cb6b5ad9df185d28eb61b7775d1b7d7587997c1f20c31ff01e45c2db9c5c89", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "CoffeeBench evaluates LLM agents as a coffee roaster in a 90-day multi-agent economy with fixed reference agents, measuring cumulative net income through autonomous communication, negotiation, and transactions.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce CoffeeBench, a benchmark for evaluating LLM agents in a long-horizon multi-agent economy composed of heterogeneous firms."}, "familyId": "bmf_43631014a768", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/SakanaAI/CoffeeBench", "url": "https://github.com/SakanaAI/CoffeeBench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/SakanaAI/CoffeeBench", "url": "https://huggingface.co/papers/2606.16613"}, {"basis": "huggingface-paper-association", "linkedUrl": "https://github.com/SakanaAI/CoffeeBench", "url": "https://huggingface.co/api/papers/2606.16613"}], "scope": "benchmark_repo", "url": "https://github.com/SakanaAI/CoffeeBench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_coffeebench_22526372", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Logistics"], "links": {"code": "https://github.com/SakanaAI/CoffeeBench", "data": null, "hfPaper": "https://huggingface.co/papers/2606.16613", "paper": "https://arxiv.org/abs/2606.16613", "pdf": "https://arxiv.org/pdf/2606.16613", "project": null, "report": "https://arxiv.org/abs/2606.16613"}, "metrics": [], "motivation": "As LLM agents become capable of increasingly long-horizon tasks, evaluating their performance in economic systems is becoming increasingly important.", "name": "CoffeeBench", "oneLine": "CoffeeBench evaluates LLM agents as a coffee roaster in a 90-day multi-agent economy with fixed reference agents, measuring cumulative net income through autonomous communication, negotiation, and transactions.", "paperTitle": "CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies", "primaryDomain": "Logistics & Operations", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.16613", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Sakana AI", "organizationType": "company-research-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/SakanaAI/CoffeeBench"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-15", "source": {"authors": ["Issa Sugiura", "Daichi Hattori", "Kazuo Araragi", "Keita Ogawa", "Shota Onose", "Taro Makino", "Teppei Usuki", "Takashi Ishida"], "categories": ["cs.AI"], "id": "2606.16613", "title": "CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies", "type": "arxiv", "url": "https://arxiv.org/abs/2606.16613"}, "sourceUpdatedAt": "2026-06-15T12:04:44Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Existing benchmarks often focus on single agents in static environments, whereas CoffeeBench captures long-horizon multi-agent economic interactions. It provides a scoring contract based on net income, enabling comparison of agent economic decision-making and revealing failure modes like idle-drift."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
