T3 Code claims 2x OpenAI's token throughput — mocked for conflating routing with inference
scaling01 · x · 2026-11-02
theo claimed T3 Code does roughly 2x the token throughput of all of OpenAI last year. scaling01 pushed back sarcastically: that's comparing inference throughput with token routing — the figures measure different things, making the claim apples-to-oranges.
More from Infra
- Gary Marcus flags CoreWeave's 57.88% six-month stock collapse ahead of Nov 10 earnings — GaryMarcus · 2026-11-03
- Pinokio ships "Remote AI" update for controlling local AI from anywhere — cocktailpeanut · 2026-11-03
- Pinokio Update Brings "Remote AI": Control Local AI Apps From Anywhere — cocktailpeanut · 2026-11-03
- Pinokio launches Remote: bring locally installed AI apps anywhere — cocktailpeanut · 2026-11-03
- Pinokio Remote Launches: Access Your Local AI Apps From Any Device — cocktailpeanut · 2026-11-03
- Wall Street's new narrative: hyperscalers turn cash flow positive in late 2027 — inductionheads · 2026-11-03