Leaked BenchCAD data shows OpenAI's internal model costs up to $37 per task, fueling $500 plan speculation
imjustnewatai · x · 2026-11-04
Cost-per-task vs. score data on BenchCAD suggests OpenAI's internal model runs $11/task at 76%, $23 at 83%, $34 at 86%, and $37 at 88% — with a max tier not shown. For comparison, Astra's max tier costs $11/task and scores 59%.
The poster argues this explains OpenAI's $500/month plan: the model is "expensive as fuck" and very fat, with speculation of roughly 50 trillion parameters. Data is extrapolated by Astra and unverified.
Related event: Leaked BenchCAD Slides Show OpenAI Internal Model Far Ahead of GPT-6 Astra(4 posts)→
More from Infra
- AMD exec gifts DHH a 192GB AMD workstation, embraces Omarchy Linux — AnushElangovan · 2026-11-04
- How LLM inference actually works: from 140GB weights to GPUs and model serving, explained as a story — _jaydeepkarale · 2026-11-04
- Running 4x RTX 3090 on a consumer board: a Qwen3.8 local inference build — Civil_Fee_7862 · 2026-11-04
- LlamaIndex speeds up doc-parsing model OCRia-1 10x to 2800 tok/s on one B200 — llama_index · 2026-11-04
- Coding agents more than double every released speculative decoding system, DFlash2 hits 10.87x — berkeley_ai · 2026-11-04
- Scaling Python Without a Dockerfile: Metaflow, DuckDB and Parquet in Practice — anacondainc · 2026-11-04