WeirdML v3: Claude Haiku 5.5 beats all non-Anthropic/OpenAI models by a wide margin
scaling01 · x · 2026-10-14
New WeirdML v3 benchmark results show Claude Haiku 5.5 beating all non-Anthropic/OpenAI models by a good margin, while Mistral Large 4 and Inkling underperform (both run via opencode), with GPT 5.2 also in the comparison. Muse and Argon runs are still in progress.
More from Models
- GPT-6 Astra Reads Only 46% of VINs Correctly vs Sonnet 5's 86%: Routing Pays Off — arnabdotorg · 2026-10-15
- VLMs May Turn Clearly Superhuman in Vision Within a Year or Two, Researcher Predicts — arnabdotorg · 2026-10-15
- Gemini 4 Argon Trails GPT 6 Astra by Just 25 Points on Physical Agents Index — qinzytech · 2026-10-15
- GPT-6-Astra scores 98% on unpublished theoretical physics problems, catching expert errors — sandersted · 2026-10-15
- Devs report quality degradation when Kimi K3 is served via third-party providers — casper_hansen_ · 2026-10-15
- Gartner's 2026 GenAI Hype Cycle: Agentic AI still climbing, AGI 10+ years out — luisdans · 2026-10-15