Open Source Factory

IMAGE, VIDEO & AUDIO Sol-H3: Keep the Eight-GPU Setup Beside the Timing Open Source Factory · Sep 07, 2026 · 4 min read NVIDIA's 1.653-second figure uses an eight-B300 system and a four-step profile. It is a specific result for Sol-H3's serving setup. Removing the hardware or step count from the sentence would make it a poor guide to the time a desktop user should expect. In this article The Sol-H3 runtime Timings for the specified profiles Different H3 deployment paths Quality, license and hardware limits Linked source documents One paragraph of context before the numbers. Sol-H3 is not a new model. It is NVIDIA's Sol-Engine and Sol-Attn runtime wrapped around MiniMax's open-weight H3, running a 4-step profile instead of the 50-step base. The weights are still MiniMax's under their community license; the stack code is Apache 2.0. And the headline 11x mixes two things, fewer steps plus a faster engine, which NVIDIA states openly. Keep that split in mind and t...

IMAGE, VIDEO & AUDIO H3 Max: The Hosted Retrain and the Open Base Open Source Factory · Aug 28, 2026 · 5 min read fal's H3 Max is a post-trained H3 service, not the same deployment as the downloadable base model. Its reported speed and arena placement make an API comparison worth considering. Local editing, released artifacts and hosting requirements need a separate check. In this article The base model and fal's post-training Output examples Latency claims and serving conditions The dated arena rankings Price and launch conditions Generation, editing and local use H3 Max uses the 33B H3 base, with post-training aimed at prompt adherence and visual quality. At the article's leaderboard snapshot, Artificial Analysis placed it first in image-to-video and third in text-to-video with audio. Its 768p limit and unreleased weights distinguish it from the base H3's local and editing workflows. From open weights to a #1-ranked retrain in about a month (author's mile...

IMAGE, VIDEO & AUDIO LTX-2.5: Reproduce the Setup Before Quoting the Speed Open Source Factory · Aug 11, 2026 · 7 min read LTX's 6.8-second result uses two GB200 accelerators for a specified 720p clip. The open weights and ComfyUI integration make other setups possible, not equally fast. For a local trial, keep the artifact, resolution, frame count and complete pipeline time in the record. In this article Architecture and released variants The 6.8-second test setup LTX's blind preference test Documented local hardware paths License conditions Earlier open-video releases World-model and real-time plans The August 11 release generates video and audio together. It offers Hugging Face weights, launch-day ComfyUI integration and a hosted API. Each path has different hardware, pricing and reproducibility conditions. That day-one ComfyUI integration is the part that actually matters. ComfyUI has quietly become the de facto prototyping environment for open generative media, and...

IMAGE, VIDEO & AUDIO MiniMax H3's Early Local Runs: Settings Before Speed Open Source Factory · Aug 04, 2026 · 6 min read AI Video Open Weight Early H3 reports mix different GPUs, resolutions and step counts. They are useful for finding configurations to try, but do not form a controlled speed ranking. This roundup keeps those reports separate from the narrow 1080 Ti observation and the hosted service's capabilities. In this article Model and service specifications Early community impressions Reported GPU timings and settings Output strengths and limitations Launch API prices Weight-license conditions Local artifacts and workflow What the early reports establish Linked source documents This roundup records community reports from the first three days after release: GPU timings, output impressions and the license terms that accompanied the weights. Those reports use different prompts and settings, so the timing table is not a controlled GPU benchmark. Model and servic...