Report shows shifts in AI model popularity across text, image, video

Poe, a platform for exploring and comparing AI models, has released its “Early 2025 AI Ecosystem Trends” report revealing significant shifts in user preferences across text, image, and video generation models. According to the report, OpenAI and Anthropic dominate text generation with approximately 85% of message share, while newcomers like DeepSeek and Google’s Gemini are gaining traction. Since Claude 3.5 Sonnet’s launch in June 2024, usage across OpenAI and Anthropic models has become nearly equal among Poe subscribers.

In image generation, BlackForestLabs’ Flux family has emerged as the frontrunner with close to 40% of messages, while Google’s Imagen3 has grown steadily to capture almost 30% of usage share. Early pioneers like Dall-E 3 and Stable Diffusion have seen their relative usage drop by nearly 80% as the field has expanded from 3 to approximately 25 official models.

The video generation category, which only began in late 2024, has seen Google’s Veo 2 quickly capture nearly 40% of messages, challenging incumbent Runway which maintains 30-50% of video generation messages despite offering only a single model since September 2024.

Poe’s report also notes that as frontier labs release more capable models, users quickly migrate to newer versions, with older models seeing rapid declines in usage. DeepSeek models notably achieved higher usage levels than previous open-source model families like Llama and Mistral, reaching 7% of messages at their peak despite only launching in December 2024.

via VentureBeat

Related posts:

Stay up-to-date: