Stuubzzz Creative Studios blog https://stuubzzz.studio/blog/ Settings, measured timings and teardowns for self-hosted AI video in ComfyUI. en Thu, 08 Oct 2026 08:00:00 GMT MiniMax H3 timeline in ComfyUI: middle frames, extending a clip, pinned sounds and clips https://stuubzzz.studio/blog/minimax-h3-keyframes-extend-video-timeline-comfyui/ https://stuubzzz.studio/blog/minimax-h3-keyframes-extend-video-timeline-comfyui/ Thu, 08 Oct 2026 08:00:00 GMT Ultimate H3 0.11 lets you direct MiniMax H3 in ComfyUI with up to two middle frames, a clip continued from its last 1.6 s, and up to 8 pinned sounds or clips. ComfyUI errors beginners hit: what each of the eight means and how to fix it https://stuubzzz.studio/blog/comfyui-errors-how-to-fix/ https://stuubzzz.studio/blog/comfyui-errors-how-to-fix/ Thu, 08 Oct 2026 08:00:00 GMT The free Part 1 of ComfyUI From Zero covers eight errors beginners actually see, from Value not in list to CUDA out of memory: each message, meaning and fix. MiniMax H3 in 4 steps: PDMD LoRA vs NVIDIA's SoL Refiner, timed on an RTX 5090 https://stuubzzz.studio/blog/minimax-h3-4-steps-pdmd-vs-sol-refiner/ https://stuubzzz.studio/blog/minimax-h3-4-steps-pdmd-vs-sol-refiner/ Thu, 08 Oct 2026 08:00:00 GMT PDMD's 4-step LoRA finished a 6.6 s MiniMax H3 clip to 2520×1440 in 101 s on an RTX 5090, against 753 s for 20 steps plus polish. SoL Refiner changed faces. LTX 2.5 modes in ComfyUI: what each one does and how to stop MSR duplicate characters https://stuubzzz.studio/blog/ltx-2-5-comfyui-modes-msr-duplicate-characters/ https://stuubzzz.studio/blog/ltx-2-5-comfyui-modes-msr-duplicate-characters/ Thu, 08 Oct 2026 08:00:00 GMT Each LTX 2.5 mode in ComfyUI wants its own prompt. With MSR V1, naming @ref1 in every action cut seeds with a duplicate character from 2 of 4 to 0 of 4. MiniMax H3 ComfyUI workflow: Speed, Hybrid and Quality modes timed on an RTX 5090 https://stuubzzz.studio/blog/stubelius-ultimate-h3-minimax-h3-comfyui-workflow/ https://stuubzzz.studio/blog/stubelius-ultimate-h3-minimax-h3-comfyui-workflow/ Wed, 30 Sep 2026 08:00:00 GMT Stubelius Ultimate H3, a free MiniMax H3 workflow for ComfyUI: three presets, 8 s clip times and peak VRAM on an RTX 5090, seed hunting and install. LTX 2.5 blurry in ComfyUI: render the first pass at full size https://stuubzzz.studio/blog/ltx-2-5-blurry-full-size-first-pass/ https://stuubzzz.studio/blog/ltx-2-5-blurry-full-size-first-pass/ Wed, 30 Sep 2026 08:00:00 GMT Soft LTX 2.5 video often comes from a half-size first pass. Same seed on an RTX 5090: 61.1 s at full size vs 46.2 s at 0.5 plus a x2 latent refine. MiniMax H3 sampler and scheduler settings: three safe pairs, two that break https://stuubzzz.studio/blog/minimax-h3-sampler-scheduler-settings/ https://stuubzzz.studio/blog/minimax-h3-sampler-scheduler-settings/ Wed, 30 Sep 2026 08:00:00 GMT Five MiniMax H3 workflows at 6 to 14 steps: euler/simple, euler/beta and er_sde/beta had no gross failure. dpmpp_2m_sde/karras gave noise on four of five. LTX 2.5 VAE decode: flicker every 56 frames and the shared-memory stall on Windows https://stuubzzz.studio/blog/ltx-2-5-vae-decode-flicker-vram-windows/ https://stuubzzz.studio/blog/ltx-2-5-vae-decode-flicker-vram-windows/ Wed, 30 Sep 2026 08:00:00 GMT LTX 2.5 flickers every 56 frames when VAE Decode (Tiled) uses temporal overlap 8. On Windows an oversized decode spills into shared GPU memory. Two fixes. DLSS 5 vs RTX Video Super Resolution in ComfyUI: upscale times and keeping the colours https://stuubzzz.studio/blog/dlss-5-vs-rtx-vsr-comfyui-video-upscale/ https://stuubzzz.studio/blog/dlss-5-vs-rtx-vsr-comfyui-video-upscale/ Wed, 30 Sep 2026 08:00:00 GMT Upscaling an 8 s MiniMax H3 clip to 1440p in ComfyUI on an RTX 5090: RTX VSR 1:06, DLSS 5 + Color Lock 4:27. Plus LTX 2.5 times and the colour fix. MiniMax H3 prompt format: three fields, speaker tags and a word budget https://stuubzzz.studio/blog/minimax-h3-prompt-format-three-fields/ https://stuubzzz.studio/blog/minimax-h3-prompt-format-three-fields/ Wed, 30 Sep 2026 08:00:00 GMT The structured prompt MiniMax H3 expects: alignment line, three named fields, (S1) speaker IDs, dialogue tags, shot timestamps and the 17k+5 frame grid. MiniMax H3 TaoMate LoRA, Singularity and turbo: which pairings hold colour https://stuubzzz.studio/blog/minimax-h3-taomate-singularity-turbo-lora-pairing/ https://stuubzzz.studio/blog/minimax-h3-taomate-singularity-turbo-lora-pairing/ Wed, 30 Sep 2026 08:00:00 GMT Eight MiniMax H3 runs on one seed: TaoMate 3-step on Singularity v1.3 drops saturation from 42.8 to 28.6. TaoMate on stock and turbo on both hold colour. AI Toolkit MiniMax H3 LoRA training: what the Distillation Handling Method setting does https://stuubzzz.studio/blog/ai-toolkit-minimax-h3-lora-distillation-handling/ https://stuubzzz.studio/blog/ai-toolkit-minimax-h3-lora-distillation-handling/ Wed, 30 Sep 2026 08:00:00 GMT From the AI Toolkit code at commit ecee894: what Training Adapter and Contrastive Guidance set for MiniMax H3 LoRAs, and when Differential Guidance runs. Stubelius Ultimate LTX 2.5 ComfyUI workflow: multi-shot timeline, seed hunt, Licon MSR https://stuubzzz.studio/blog/ltx-2-5-director-comfyui-multishot-timeline/ https://stuubzzz.studio/blog/ltx-2-5-director-comfyui-multishot-timeline/ Sat, 12 Sep 2026 08:00:00 GMT Free LTX 2.5 ComfyUI workflow: multi-shot timeline, 1 to 4 seeds with sound, Licon MSR references, up to 4K. Install steps and RTX 5090 timings. MiniMax H3 Director in ComfyUI: seed hunt, two-stage sampling and the Stubelius refine https://stuubzzz.studio/blog/minimax-h3-director-seed-hunt-two-stage-comfyui/ https://stuubzzz.studio/blog/minimax-h3-director-seed-hunt-two-stage-comfyui/ Sat, 12 Sep 2026 08:00:00 GMT Stubelius Director for MiniMax H3 in ComfyUI: scout up to 4 seeds at 0.25 MP, finish one with a learned 2x latent upscale. Settings and install. FastH3 vs Turbo LoRA on MiniMax H3: 5.1 vs 6.21 minutes per clip on an RTX 5090 https://stuubzzz.studio/blog/fasth3-vs-turbo-lora-minimax-h3-speed-test/ https://stuubzzz.studio/blog/fasth3-vs-turbo-lora-minimax-h3-speed-test/ Sat, 12 Sep 2026 08:00:00 GMT FastH3 preview adapter at 6 steps vs the 8-step Turbo LoRA on MiniMax H3: 5.1 vs 6.21 min for a 15 s clip on an RTX 5090, same time per step. FastH3 LoRA does nothing in ComfyUI: why, and the converter that fixes it https://stuubzzz.studio/blog/fasth3-lora-comfyui-converter/ https://stuubzzz.studio/blog/fasth3-lora-comfyui-converter/ Sat, 12 Sep 2026 08:00:00 GMT FastVideo's FastH3 preview adapter loads in ComfyUI and changes nothing: 0 of 389 module names match the pruned MiniMax H3. Key map, converter, settings. LTX 2.5 sampler settings: euler_ancestral or res_2s, and when you need the distilled LoRA https://stuubzzz.studio/blog/ltx-2-5-samplers-euler-vs-res-2s-distilled-lora/ https://stuubzzz.studio/blog/ltx-2-5-samplers-euler-vs-res-2s-distilled-lora/ Sat, 12 Sep 2026 08:00:00 GMT LTX 2.5 distilled in ComfyUI: euler_ancestral + linear_quadratic at 8 steps, cfg 1, matches Lightricks' sigmas. Plus when the distilled LoRA is needed. MiniMax H3 2.5D anime style LoRA: what the retrain changed and which strength to use https://stuubzzz.studio/blog/minimax-h3-2-5d-anime-style-lora/ https://stuubzzz.studio/blog/minimax-h3-2-5d-anime-style-lora/ Sat, 12 Sep 2026 08:00:00 GMT MiniMax H3 slides to flat anime or 3D. A style LoRA retrained on 50 three-second clips holds a glossy 2.5D look. Same-seed strength sweep from 0.4 to 1.0. Consistent character LoRA dataset: the five views and the rules https://stuubzzz.studio/blog/consistent-character-dataset-lora-training/ https://stuubzzz.studio/blog/consistent-character-dataset-lora-training/ Sat, 12 Sep 2026 08:00:00 GMT The 15 to 30 images for a character consistency LoRA: front, side, 3/4, back and face detail, what to lock, how to caption, and what changes for video.