Rant: Gaming AI Model Benchmarks with Custom Metrics
JJitsev · x · 2026-07-13
The author rants about the current chaos in AI model evaluation: making one's own model look the strongest by inventing scoring metrics that no one else uses. For example, in the SOOFI evaluation, using a so-called "capability index" to evaluate base models made Qwen 3 32B appear weaker than practically insignificant models like Teuken.
Related event: Controversy Over Nemotron Reproduction and Custom Benchmarks(4 posts)→
More from Models
- Fern's "context is intelligence" thesis: prediction keeps improving out to 10.2M tokens — bneyshabur · 2026-10-26
- Mirendil launches Fern, a from-scratch pretrained model with up to 10M-token context — bneyshabur · 2026-10-26
- Users spot a dedicated translation UI appearing in Claude — BLCNYY · 2026-10-26
- GrokBot user: flight booking and shopping all blocked, agent can't do anything useful — lucasmeijer · 2026-10-26
- Vending-bench-style eval: Sol 5.6 appears the most aggressive contract-violator — xuanalogue · 2026-10-26
- Fable 5.5 launch rumors persist for third week, but no frontier model drops this month — patience_cave · 2026-10-26