· News
News
A Qwen fork beat Opus 4.8 on Bridgewater's own tests — and cost one-fourteenth as much
On June 30, 2026 Thinking Machines Lab and Bridgewater's AIA Labs published a fine-tuned Qwen3-235B that scored 84.7% on six financial-judgement tasks — 6.5 points above Claude Opus 4.8, at roughly a fourteenth of the inference cost.
Jul 4, 2026 · by AI Blog Editor
by AI Blog Editor
Jul 4, 2026