GPT 5.5 + Claude 4.8 Ensemble Powered AIRA₃'s 8th-Ranked Gold Medal Kaggle Entry
AIatMeta · x · 2026-09-06
Meta details the model combinations behind AIRA₃'s live Kaggle run: the 8th-ranked gold medal entry was an ensemble of GPT 5.5 (with OpenCode) + Claude 4.8 (with ClaudeCode).
Post-hoc submissions, graded externally on the same private test set, showed Muse Spark 1.2 (with MuseCode) also hitting gold, while Muse Spark 1.1 and GLM 5.2 (both with OpenCode) reached silver — suggesting the system is robust across underlying models.
Related event: Meta's AIRA₃ Autonomous Research System Wins Kaggle Gold(5 posts)→
More from coding & agent
- Microsoft's FrontierTour: AI governance becomes everyone's problem as agents take action — donasarkar · 2026-10-23
- Grady Booch revives Weinberg's 1971 quip: software so buggy one woodpecker could destroy civilization — Grady_Booch · 2026-10-23
- One Agent Replaces Half an SDR Team: Full Lead-Research-to-Email Workflow — tibo_maker · 2026-10-23
- Why don't coding agent harnesses ship with their own sandbox? Only amp does — Aryvyo · 2026-10-23
- Dev builds a VST synth with Codex and MCP, lets the AI jam in his studio — Yamapama · 2026-10-23
- AgentsInTheCloud update: use your Claude subscription with built-in coding agents — lucasmeijer · 2026-10-23