AnimateLCM-SVD-Comfy
Kijai基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基本情報を収録・詳細は順次追加
言語から画像、オープンウェイトからAPIまで。能力と実際の費用を比較。
次のアイデアに合うAIを見つけましょう。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基本情報を収録・詳細は順次追加
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基本情報を収録・詳細は順次追加
`flux1-dev-fp8.safetensors` is `float8_e4m3fn`, original upload name kept for backwards compatibility. https://huggingface.co/Shakker-Labs/FLUX.1-dev-ControlNet-Union-Pro flux1-dev-fp8.safetensors, flux_shakker_labs_union_pro-fp8_e4m3fn and it's variants falls under the [`FLUX.1 [dev]` Non-Commercial License](https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md). flux1-schnell-fp8.safetensors falls under the Apache-2.0 License.
翻訳待ち・原文を表示
To be used with ComfyUI native HunyuanVideo implementation, or my wrapper: https://github.com/kijai/ComfyUI-HunyuanVideoWrapper GGUF's created using city96's scripts, and only works with their nodes: https://github.com/city96/ComfyUI-GGUF
翻訳待ち・原文を表示
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基本情報を収録・詳細は順次追加
The fp8 quantizations were done with the basic static weight scales and are set to not run with fp8 matmuls, the models marked `input_scaled` additionally have activation scaling, and are set to run with fp8 matmuls on supported hardware (roughly 40xx and later Nvidia GPUs). As this is first time I'm attempting to do calibrate input scales, these are pretty experimental, but result wise seems to work, this is a test on a 4090, 8 steps with distill: **update:** input_scaled_v3 follows the same pattern as the official one where blocks 0-1 and 46-47 (first two and last two) are kept in bf16, as well as better calibrated input scales, this does fix some of the issues in v2, especially when using input audio.
翻訳待ち・原文を表示
Starting from this commit https://github.com/Comfy-Org/ComfyUI/commit/f266b8d352607799afb4adf339cdfa854025185e Embedding connector has been moved from text encoder to diffusion model, so to continue using single files you now have to update KJNodes and use my loaders: LTX2 Tiny VAE has been trained by madebyollin, and now has preliminary support in ComfyUI nightly version.
翻訳待ち・原文を表示
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基本情報を収録・詳細は順次追加
Can be used with: https://github.com/kijai/ComfyUI-WanVideoWrapper and ComfyUI native WanVideo nodes. I've also started to do fp8_scaled versions over here: https://huggingface.co/Kijai/WanVideo_comfy_fp8_scaled SkyReels: https://huggingface.co/collections/Skywork/skyreels-v2-6801b1b93df627d441d0d0d9
翻訳待ち・原文を表示
Better fp8 scaled models (when measured against fp16) based on quantization code from https://github.com/Tencent-Hunyuan/HunyuanVideo/blob/main/hyvideo/modules/fp8_optimization.py Can be used with: https://github.com/kijai/ComfyUI-WanVideoWrapper (latest version) and ComfyUI native WanVideo nodes. The e5m2 marked as v2 is the one uploaded here and these are all scaled even if I forgot to label properly.
翻訳待ち・原文を表示
Various GGUF quants of WanVideo models and modules, mainly for ComfyUI usage. --- GGUF VACE modules can be used in the WanVideoWrapper as extra_models, or in native workflows using `GGUFLoaderKJ` node from KJNodes.
翻訳待ち・原文を表示
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
提供元・分類・入出力の条件を変更してください。新しいモデルを随時追加しています。
月間トークン数を調整してAPI費用を比較。
収集済みの比較可能な100万トークン当たり米ドル料金です。キャッシュ割引、段階料金、追加推論出力、税金は含みません。最新料金は出典で確認してください。
入力・出力・キャッシュ料金を分け、同じ用途と使用量で総費用を比較します。画像・音声・動画は枚数、時間、解像度で課金される場合があります。
必ずしもそうではありません。バージョンごとのライセンス、商用制限、ポリシーを確認してください。自前運用には計算資源と保守費用も必要です。
順位はテスト、バージョン、評価方法によります。実際の業務データで精度、遅延、安定性、費用を比較してください。