Two ComfyUI workflows for LTX-2.5 that add first-frame/last-frame keyframe conditioning and audio-driven lipsync to the stock image-to-video template. The official 2.5 I2V workflow only conditions on a starting image and lets the model improvise where the clip ends up; these versions lock in an ending image too, so the motion travels from your chosen start frame to your chosen end frame, while your own audio clip drives the speech/singing.Built on top of Comfy's own LTX-2.5 template nodes. No custom node packs required. WHICH ONE SHOULD SHOULD YOU USE?SinglePass_HQ (LTX2.5_FLF2V_Lipsync_SinglePass_HQ.json)<ul><li>One full-resolution pass</li><li>Best quality, noticeably cleaner</li><li>Slower: about 20 min for a 28 s clip on my machine</li><li>Based on Comfy’s official LTX-2.5 FLF2V template, plus lipsync</li><li>Extra model: prompt enhancer (only if you turn that optional toggle on)</li></ul>TwoStage_Fast (LTX2.5_FLF2V_Lipsync_TwoStage_Fast.json)<ul><li>Low-res pass, then a 2x latent upscale and refine pass</li><li>Good quality, slightly softer</li><li>Faster: under 10 min for the same clip</li><li>Based on my original lipsync workflow, repaired</li><li>Extra model: latent spatial upscaler</li></ul>Rule of thumb: use TwoStage_Fast for drafts, long clips, or lower VRAM, and SinglePass_HQ for the final render.Usage<ol><li>Load the workflow into ComfyUI.</li><li>In Load First Frame, upload your starting image.</li><li>In Load Last Frame, upload your ending image.</li><li>In Load Driving Audio, pick the audio clip (speech or song) you want the character to perform.</li><li>Write your prompt describing the scene and motion. For lipsync, say who is speaking or singing, and include the spoken or sung words.</li><li>Set duration (seconds). It must be no longer than your audio, counted from <code>audio_start</code>.</li><li>Pick the aspect ratio and size in the Resolution Selector node.</li><li>Queue the prompt.</li></ol>REQUIREMENTS<ul><li>ComfyUI with the LTX-2.5 nodes (comfy-core, no extra custom nodes)</li><li>LTX-2.5 model files as used by the stock LTX-2.5 template:<ul><li>distilled transformer: ltx-2.5-22b-distilled-transformer-comfy-int8-convrot</li><li>video VAE and audio VAE: ltx-2.5-video-vae-bf16, ltx-2.5-audio-vae-bf16</li><li>Gemma 4 12B text encoder: gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot</li><li>TwoStage_Fast only: latent spatial upscaler, ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0</li><li>SinglePass_HQ only: prompt enhancer model, gemma4_e2b_it_int8_convrot (only used if you turn the optional prompt enhance toggle on)</li></ul></li></ul>CreditAdapted from Comfy's stock LTX-2.5 image-to-video and first/last-frame templates, using <code>LTXVAddGuide</code>, the same keyframe-conditioning node used in Lightricks' official first-frame/last-frame example workflows. Lipsync uses the LTX audio VAE encode path with a zero noise mask to hold the driving audio fixed.

Files with SHA256: aeb1d77610459a8987e2676b0bfafa4a38c10a79dffd4f4b429c5a0f4c3af987