Impressive Qwen Max Vision: Few-Shot Prompting Auto-Detects Objects
NielsRogge · x · 2026-08-11
Developers have found that Qwen 2.5 Max demonstrates impressive vision capabilities.
Using few-shot image prompting, a user only needs to annotate one bounding box in an image as an example, and the model can automatically identify and box out the remaining similar objects. The developer tested this on a Belgium vs. USA soccer match image, successfully prompting the model to find the remaining Belgian players on the field.
More from Models
- Muse Glimmer 30B Ships with DFlash, Boosting Inference Speed 2-4x — mervenoyann · 2026-08-11
- Grok 4.6 Drops Today: A Comparison of Supported Coding Tools — PawelHuryn · 2026-08-11
- Anthropic Security Tester Reports Bizarre Opus 5 Over-Refusal — Own-Director-5503 · 2026-08-11
- Security Researcher Mocks OpenAI's Cybersecurity Filters as Trivial to Bypass — evilsocket · 2026-08-11
- Claude Code Enterprise Pricing Hits 40x Markup for Same Tokens and Model — jakozaur · 2026-08-11
- Open-Source Dashboard Aggregates Real-Time Pricing for 150 AI Models Across 25 Providers — redouanea · 2026-08-11