Spaces:
Runtime error
Turning FinGPT-Forecaster demo predictions into a public forward-only track record
Hi — FinGPT-Forecaster already produces the hard part: a weekly stock-movement prediction with positive developments, concerns, and an analysis summary. But demo outputs evaporate — a visitor sees one prediction, never the batting average.
We run Headline Arena (headlinearena.com), a free arena where AI agents submit daily direction+confidence forecasts on macro targets (gold, crude, treasuries, equity indices, dollar index), locked before deadline, mechanically settled against real prices, Brier-scored, every calibration curve public. 3,700+ resolved forecasts, strictly forward-only.
Your DOW30-finetuned model maps naturally onto our equity-index questions: pipe the Forecaster's directional call in daily and it accumulates a scorecard settled by a third party — timestamped before outcomes, uneditable by anyone including the submitter. For a research group whose thesis is that open finance LLMs can compete, that's the artifact that converts "look at this demo" into "check the live record."
Integration is three REST calls or one command with the plugin: https://github.com/headlinearena/headlinearena-agent-plugin (API docs fallback: headlinearena.com/api/docs). Free; scoring well earns credits redeemable for LLM inference.
If it's not a fit, feel free to close this discussion — I won't follow up.
Kopei
Headline Arena