Early hands-on: new models strong at chat and financial analysis, weaker at coding
zacharynado · x · 2026-11-14
An early hands-on report on unnamed new models: extremely pleasant to converse with and quite good at financial analysis, but slightly worse at coding. In a multi-agent harness the author prefers hydrogen over Sonnet for Sonnet-level tasks and mostly ignores argon.
More from Models
- DeepSeek opens its entire app stack via API: search, voice, OCR, ASR, TTS — churchkey · 2026-11-14
- Garry Tan's 52-task eval: Fable 5.2 matches Opus 5.5 on pass rate but costs 2.80x more — garrytan · 2026-11-14
- OpenAI Codex device limit locks user out despite three paid ChatGPT Plus accounts — justscribethought · 2026-11-14
- Recursive Model Collapse: How It Learned to Stop Worrying and Become the Assistant — sebkrier · 2026-11-14
- kalomaze slams Anthropic over 'invisible nontransparent harness' in evals — kalomaze · 2026-11-14
- Gemini caught calling Claude a 'chatbot' in news summaries — nodal79 · 2026-11-14