Claude Opus 4.6, GPT-5.2 Score Only ~30% in New SWE-Atlas Benchmark

{
  "signal_id": "b93b6486-bf3f-426c-ad7e-9e3386e0fb0b",
  "signal_type": "news",
  "signal_subtype": "launches",
  "detected_at": "2026-03-05T07:53:41+00:00",
  "company": {
    "name": "Scale AI",
    "domain": "scale.com"
  },
  "data": {
    "url": "https://analyticsindiamag.com/ai-news/claude-opus-46-gpt-52-score-only-30-in-new-swe-atlas-benchmark",
    "title": "Claude Opus 4.6, GPT-5.2 Score Only ~30% in New SWE-Atlas Benchmark",
    "excerpt": "Claude Opus 4.6, GPT-5.2 score only ~30% in new SWE-Atlas benchmark.\n\nScale AI's new benchmark evaluates AI coding agents across a spectrum of professional software engineering tasks.\n\nMARCH 4, 2026, 11:52 PM\n\nScale AI has introduced SWE-Atlas, a new benchmark designed to evaluate how well AI coding agents perform real-world software engineering tasks inside complex codebases rather than simply generating code snippets.\n\n\"SWE-Atlas is a benchmark for evaluating AI coding agents across a spectrum of professional software engineering tasks,\" the company said in its announcement.\n\nThe benchmark includes three complementary leaderboards: Codebase QnA, Test Writing, and Refactoring.\n\nOf these, Codebase QnA is the first component released publicly, while the other two evaluations are expected to be introduced later.\n\nCodebase QnA focuses on testing how well AI agents understand large software systems before attempting modifications.\n\nThe dataset contains 124 tasks drawn from 11 production repositories written in Go, Python, C, and TypeScript. Agents are placed inside sandboxed Docker environments containing the repositories and must answer technical questions by exploring the codebase, executing commands, and analysing runtime behaviour.\n\nScale AI said that these tasks require running the software, tracing execution across multiple files, and synthesising findings.\n\nThe benchmark...",
    "product": "SWE-Atlas",
    "summary": "Scale AI, Inc. launches SWE-Atlas.",
    "category": "launches",
    "found_at": "2026-03-05T07:53:41Z",
    "planning": false,
    "image_url": "https://storage.googleapis.com/gpt-engineer-file-uploads/XW65pun0vjQL3xcUmVglEg4ZrsL2/social-images/social-1763560217027-Expert-Panel-discussion-768x512.webp",
    "confidence": 0.8973,
    "product_data": {
      "full_text": "SWE-Atlas",
      "fuzzy_match": true
    },
    "published_at": "2026-03-05T07:53:41Z",
    "article_sentence": "Scale AI has introduced SWE-Atlas, a new benchmark designed to evaluate how well AI coding agents perform real-world software engineering tasks inside complex codebases rather than simply generating code snippets."
  }
}

curl https://signals.autobound.ai/v1/signals/b93b6486-bf3f-426c-ad7e-9e3386e0fb0b \
  -H "X-API-KEY: $AUTOBOUND_API_KEY"

curl -X POST https://signals.autobound.ai/v1/companies/enrich \
  -H "X-API-KEY: $AUTOBOUND_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"domain":"scale.com","limit":20}'