FOUNDING OFFERSign up to unlock 35% off your first plan.

Stillmore author

Lan He

I'm Lan. Years of writing promotional and editorial work in the US, mostly for companies with big audiences and bigger expectations — where a single article might reach hundreds of thousands of readers, and where the good ones kept earning traffic long after the campaign ended.

32 published articles
Lan He

Articles by Lan He

7–12 of 32 articles

Stillmore Blog cover for Qwen4-Max, featuring purple radial star sculpture.

News

How to Use Qwen4-Max for Video Understanding and Content Planning

To use Qwen4-Max for video understanding and content planning, send raw clips, reference photos, and campaign rules to the qwen4-max model ID for structured JSON storyboards and staged tool plans. For API integration, connect through OpenAI Chat Completions, OpenAI Responses, or Anthropic Messages at your regional endpoint rather than DashScope whenever your workflow needs Qwen4 explicit caching to reuse fixed brand guidelines, because DashScope uses implicit caching only. Teams can pair these structured planning briefs with external asset catalogs or 3D layout tools before human editors cut the final footage. Since Qwen4-Max outputs planning briefs rather than finished social ads, independent businesses that prefer ready-to-post videos, images, and captions made from existing phone photos can delegate production and pre-delivery checks to Stillmore while keeping publishing in their own hands.

Lan He· 12 min read
Stillmore Blog cover for Microsoft-Decision-1.1, featuring lime glass decision network.

News

What Is Microsoft-Decision-1.1? Multimodal Routing, Calibration, and Pricing

Microsoft-Decision-1.1 is Microsoft's specialized decision model, generally available since November 17, 2026, built to select constrained software actions rather than generating open prose. Version 1.1 adds multimodal support for photos and screenshots alongside calibrated confidence scores so applications can act, split confidence, or abstain when inputs are unclear. Developers access it through Microsoft Foundry using input-token billing, as well as third-party gateways, to power high-volume routing and triage. Because a decision model only classifies inputs and picks software branches, it does not produce finished promotional visuals. An independent business still supplies product photos to a creative service like Stillmore, whose team handles production and pre-delivery checks before delivering ready-to-post images, videos, and captions for the merchant to publish.

Lan He· 14 min read
Stillmore Blog cover for Suno Visuals, featuring pink singing bird with microphone.

News

How to Use Suno Visuals for Music-Synced Clips, Lyric Videos, and Sequences

Open the Create page on web, iOS, or Android, select Visuals, attach a library song, text prompt, built-in character, or photo, choose Fast, Standard, or Max (for images), and generate clips up to 15 seconds, add Lip Sync vocal movement, or style on-screen lyrics inside the Visuals Editor to build Sequences saved in your Visuals library. Visuals uses separate credits from song balances, screens IP and audio cues, and applies pre-share watermarks. For independent business owners who already hold everyday product or storefront photos, yet lack time to write prompts, assemble timelines, and verify packaging accuracy, Stillmore's creative team can turn those photos into finished videos, images, and social captions while you choose when and where to publish.

Lan He· 14 min read
Stillmore Blog cover for Gemini Decision 1, featuring blue glass Gemini star sculpture.

News

How to Use Gemini Decision 1 for Image and Video QA Checks

To use Gemini Decision 1 (gemini-decision-1) for image and video QA checks, submit your media and brief once through the v1beta Gemini Decisions API and evaluate bounded rules in a single pass. Configure noul checks for binary yes/no gates, choice for routing labels, and score for ordered quality rubrics inside the questions block. Each check returns a confidence value to route ambiguous files to human review under input-only pricing where shared media is billed once. Because probabilistic scores only audit inputs rather than producing finished creatives, Stillmore offers independent businesses a hands-off alternative: you upload existing product photos and offer details while Stillmore's team may evaluate tools like Gemini Decision 1 alongside editing and pre-delivery review to deliver finished images, videos, and captions you publish yourself.

Lan He· 13 min read
Stillmore Blog cover for MAI Code 2, MAI Code 2 Flash, featuring glass coding braces above teal dunes.

News

How to Use MAI Code 2 and MAI Code 2 Flash for UI, SVG, and Visual Coding

To build frontend visuals efficiently, use MAI Code 2 to scaffold multi-file UIs, SVGs, charts, and 3D scenes, then switch to MAI Code 2 Flash for fast targeted edits and code explanations. Assign layout architecture to MAI Code 2 first so stylesheets and aspect ratios align across files, and delegate routine padding or label fixes to Flash once the view renders. Access both in GitHub Copilot, plus MAI Code 2 Flash in Microsoft Foundry preview. MAI Code 2 earned 64% preference over GPT-6 Sol in blind visual evaluations, though neither model targets very long-horizon agentic work. If you run an independent business and want ready-to-post images, videos, and captions made from store photos rather than code, Stillmore handles creative direction, production, and pre-delivery checks while you decide when to publish.

Lan He· 14 min read
Stillmore Blog cover for ERNIE 5.5 Flash Preview, featuring silver geometric cube.

News

How to Use ERNIE 5.5 Flash Preview for 3D Scene Blocking and Video Previs

To use ERNIE 5.5 Flash Preview for 3D scene blocking and video previs, access the text-only model through Baidu Qianfan or OpenRouter inside an agent harness to generate Blender Python (bpy) layout scripts, render an untextured preview clip, and pass that motion guide to an external video generator. Because a text model cannot view the Blender viewport, pair your scripts with automated geometry assertions, camera frustum checks, and human playback review before rendering downstream footage. Use fast untextured blockouts to test framing and shot pacing, reserving full shaders and licensed external assets for final look development. If you run an independent business and want finished social ads without maintaining 3D code, Stillmore turns your existing phone photos into checked videos, images, and captions ready for you to publish.

Lan He· 14 min read

More voices. More ideas.

Get to know the people behind the Stillmore blog.

Meet all authors ↗