Agnify Bloghttps://blog.agnify.ai/How we turn camera feeds into operational intelligence.enGPT-6 Astra takes the lead on grounded video QAhttps://blog.agnify.ai/gpt-6-astra-grounded-video-qa/https://blog.agnify.ai/gpt-6-astra-grounded-video-qa/Wed, 30 Sep 2026 00:00:00 +0000Anton GritsayEngineeringA small change to how pipeline stages refer to objects put GPT-6 Astra clearly ahead of Gemini 3.7 Flash, at 28 times the cost.Detailed comparison of Gemini and Qwen on complex visual grounding taskshttps://blog.agnify.ai/gemini-vs-qwen-grounded-video-qa/https://blog.agnify.ai/gemini-vs-qwen-grounded-video-qa/Mon, 14 Sep 2026 00:00:00 +0000Theo ChervinskiiEngineeringGemini is marginally better at video reasoning, but both are on par in grounding. We show that pipeline design still matters the most.Qwen3.8-27B is now live at Agnify.aihttps://blog.agnify.ai/qwen3-8-27b/https://blog.agnify.ai/qwen3-8-27b/Mon, 17 Aug 2026 00:00:00 +0000Argo SaakyanEngineeringThe benchmark tables, what the visual gains actually average out to, and the latency we paid for them on our own hardware.D-FINE-seg adds semantic segmentation: a DETR model benchmarked against YOLO26 and RF-DETRhttps://blog.agnify.ai/d-fine-seg-semantic-segmentation/https://blog.agnify.ai/d-fine-seg-semantic-segmentation/Fri, 17 Jul 2026 00:00:00 +0000Argo SaakyanEngineeringD-FINE-seg now supports three tasks on the same DETR-family architecture — and on Cityscapes it benchmarks better than YOLO26 and RF-DETR.Adding a new video detector, using fire as the examplehttps://blog.agnify.ai/fire-detection/https://blog.agnify.ai/fire-detection/Wed, 10 Jun 2026 00:00:00 +0000Argo SaakyanEngineeringArgo stands up a fire detector on Agnify end to end and runs it live - a quick look at how little it takes to add a new video detector.