Open-Source Ling 3.0 Flash VL Hits OpenRouter With Free Access and FP4/INT4 Weights
FellMentKE · x · 2026-09-11
InclusionAI's Ling 3.0 Flash VL — a 124B total / 5.5B active MoE with native vision and visual agent capabilities, 262K context, hybrid instant/reasoning and tool calling — is now on OpenRouter with two weeks of free access, and FP4/INT4 quantized weights are open source on Hugging Face. A demo shows it building a football highlight reel by searching the match timeline, finding goals, verifying clips and merging them.
More from Multimodal
- A Dreamina + GPT-6 Astra pipeline turns multi-angle product images into 360° interactive pages — HeyNayeem · 2026-09-12
- Open-sourced workflow turns ideas into storyboards with GPT image, then films via Seedance — alexcovo_eth · 2026-09-12
- Creator builds 3D models and relights Gaussian splats on a phone, zero clicks, all via Astra agents — bilawalsidhu · 2026-09-12
- Runway ships Dev MCP so non-developers can integrate its generative media API — tlakomy · 2026-09-12
- This extreme-backlight Nano Banana prompt cuts a silhouette on pure white — techhalla · 2026-09-12
- A poetic eight-phrase Chinese image prompt is going viral for its classical mood control — sven_ai · 2026-09-12