OpenAI’s GPT-Image-2 still leads, while Seedance 2.5 may reset video generation
mark_k · x · 2026-07-22
- The author says no image model has yet clearly unseated GPT-Image-2 in overall quality and reliability.
- After testing Qwen Image 3.0, they see it as a meaningful improvement over the previous version, but still far behind OpenAI’s model.
- By contrast, video generation is progressing faster: Seedance 2.5 is described as potentially outperforming everything that came before it.
- The poster argues the difference may be economic: image generation has a smaller commercial market, while video has a much larger one and attracts more investment.
More from Models
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11
- User Hails Unconfirmed 'DeepSeek 4.1 Flash' as an Inflection Point in LLMs — himanshustwts · 2026-09-11
- Terminal Bench v4: GLM-5.3 Leads at 41.9%, Kimi-K3 Underwhelms at 12.6% — Ok_Warning2146 · 2026-09-11
- GPT-6 Astra beats Factorio with enemies in 44 in-game hours at ~$4,500 API cost — liminal_bardo · 2026-09-11