{"id":2740,"date":"2026-09-19T20:13:17","date_gmt":"2026-09-19T20:13:17","guid":{"rendered":"https:\/\/convly.ai\/?p=2740"},"modified":"2026-09-19T20:13:17","modified_gmt":"2026-09-19T20:13:17","slug":"ae-safetensors","status":"publish","type":"post","link":"https:\/\/convly.ai\/pt\/ae-safetensors\/","title":{"rendered":"AE Safetensors: O que \u00e9 ae.safetensors e como us\u00e1-lo em fluxos de trabalho FLUX"},"content":{"rendered":"<div class=\"convly-tldr\">\n<ul>\n<li><strong>ae.safetensors<\/strong> is the VAE (variational autoencoder) weight file shipped with Black Forest Labs&#8217; FLUX.1 image models. It decodes FLUX&#8217;s latents into RGB pixels and encodes images back to latents for img2img.<\/li>\n<li>Download it from the official <a href=\"https:\/\/huggingface.co\/black-forest-labs\/FLUX.1-schnell\/blob\/main\/ae.safetensors\" rel=\"noopener\" target=\"_blank\">black-forest-labs\/FLUX.1-schnell<\/a> or <a href=\"https:\/\/huggingface.co\/black-forest-labs\/FLUX.1-dev\/blob\/main\/ae.safetensors\" rel=\"noopener\" target=\"_blank\">FLUX.1-dev<\/a> repos on Hugging Face. The file is ~335 MB.<\/li>\n<li>In ComfyUI place it in <code>ComfyUI\/models\/vae\/<\/code> and load it with the <em>Load VAE<\/em> node. In Forge\/A1111 place it in <code>models\/VAE\/<\/code>.<\/li>\n<li>The same <code>ae.safetensors<\/code> works for FLUX.1-dev, FLUX.1-schnell and most FLUX finetunes \u2014 you do not need a separate VAE per checkpoint.<\/li>\n<\/ul>\n<\/div>\n<p><strong>ae.safetensors<\/strong> is the autoencoder (VAE) weights file that ships alongside Black Forest Labs&#8217; FLUX.1 diffusion models. FLUX generates images in a compressed latent space; <code>ae.safetensors<\/code> is the decoder that turns those latents into visible pixels (and the encoder used for image-to-image and inpainting). Without it, ComfyUI, Forge, SwarmUI or diffusers cannot render a final image from a FLUX checkpoint.<\/p>\n<p>The file is stored in the <a href=\"https:\/\/github.com\/huggingface\/safetensors\" rel=\"noopener\" target=\"_blank\">safetensors<\/a> format \u2014 a memory-mappable, tensor-only container that cannot execute arbitrary Python on load, unlike legacy <code>.ckpt<\/code> pickles. That is why nearly every modern model release ships weights as <code>.safetensors<\/code>.<\/p>\n<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_87_1 counter-flat ez-toc-counter ez-toc-container-direction\">\n<label for=\"ez-toc-cssicon-toggle-item-6aaf1a9c0568f\" class=\"ez-toc-cssicon-toggle-label\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #000000;color:#000000\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #000000;color:#000000\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/label><input type=\"checkbox\"  id=\"ez-toc-cssicon-toggle-item-6aaf1a9c0568f\"  aria-label=\"Toggle\" \/><nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/convly.ai\/pt\/ae-safetensors\/#What_Exactly_Is_in_aesafetensors\" >What Exactly Is in ae.safetensors<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/convly.ai\/pt\/ae-safetensors\/#Where_to_Download_aesafetensors\" >Where to Download ae.safetensors<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/convly.ai\/pt\/ae-safetensors\/#Installing_aesafetensors_by_Tool\" >Installing ae.safetensors by Tool<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/convly.ai\/pt\/ae-safetensors\/#VRAM_and_Performance_Notes\" >VRAM and Performance Notes<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"https:\/\/convly.ai\/pt\/ae-safetensors\/#Common_aesafetensors_Errors\" >Common ae.safetensors Errors<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"https:\/\/convly.ai\/pt\/ae-safetensors\/#aesafetensors_vs_the_All-in-One_FLUX_Checkpoint\" >ae.safetensors vs the All-in-One FLUX Checkpoint<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"https:\/\/convly.ai\/pt\/ae-safetensors\/#Where_FLUX_Fits_in_the_Broader_Model_Landscape\" >Where FLUX Fits in the Broader Model Landscape<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-8\" href=\"https:\/\/convly.ai\/pt\/ae-safetensors\/#Frequently_Asked_Questions\" >Frequently Asked Questions<\/a><\/li><\/ul><\/nav><\/div>\n<h2><span class=\"ez-toc-section\" id=\"What_Exactly_Is_in_aesafetensors\"><\/span>What Exactly Is in ae.safetensors<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>FLUX.1 is a rectified-flow transformer that operates on 16-channel latents at 1\/8 spatial resolution. The <code>ae.safetensors<\/code> file contains only the VAE weights \u2014 no transformer, no text encoders. Approximate specs:<\/p>\n<table>\n<thead>\n<tr>\n<th>Property<\/th>\n<th>Value<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>File size<\/td>\n<td>~335 MB (fp16)<\/td>\n<\/tr>\n<tr>\n<td>Latent channels<\/td>\n<td>16<\/td>\n<\/tr>\n<tr>\n<td>Spatial compression<\/td>\n<td>8\u00d7 (a 1024\u00d71024 image \u2192 128\u00d7128\u00d716 latent)<\/td>\n<\/tr>\n<tr>\n<td>Format<\/td>\n<td>safetensors (tensor-only, no pickle)<\/td>\n<\/tr>\n<tr>\n<td>Compatible checkpoints<\/td>\n<td>FLUX.1-dev, FLUX.1-schnell, FLUX.1-pro (API), most FLUX finetunes and LoRAs<\/td>\n<\/tr>\n<tr>\n<td>License<\/td>\n<td>Non-commercial for FLUX.1-dev VAE; Apache-2.0 for FLUX.1-schnell VAE<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p>The FLUX VAE is not the same file as the SD 1.5 or SDXL VAEs. Its 16-channel latent space is incompatible with those models \u2014 do not swap them.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"Where_to_Download_aesafetensors\"><\/span>Where to Download ae.safetensors<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Get it from Black Forest Labs&#8217; official Hugging Face repos. Both files are the same architecture; the schnell repo is Apache-2.0 licensed and freely redistributable, which is why most third-party bundles link to it.<\/p>\n<ul>\n<li><a href=\"https:\/\/huggingface.co\/black-forest-labs\/FLUX.1-schnell\/blob\/main\/ae.safetensors\" rel=\"noopener\" target=\"_blank\">black-forest-labs\/FLUX.1-schnell \u2192 ae.safetensors<\/a> (Apache-2.0)<\/li>\n<li><a href=\"https:\/\/huggingface.co\/black-forest-labs\/FLUX.1-dev\/blob\/main\/ae.safetensors\" rel=\"noopener\" target=\"_blank\">black-forest-labs\/FLUX.1-dev \u2192 ae.safetensors<\/a> (FLUX.1-dev non-commercial license; requires accepting the gate)<\/li>\n<\/ul>\n<p>Command-line download with the Hugging Face CLI:<\/p>\n<pre><code>pip install -U \"huggingface_hub[cli]\"\nhuggingface-cli login   # only needed for gated FLUX.1-dev\nhuggingface-cli download black-forest-labs\/FLUX.1-schnell ae.safetensors --local-dir .<\/code><\/pre>\n<p>Verify the SHA-256 against the hash shown on the file&#8217;s Hugging Face page before using it.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"Installing_aesafetensors_by_Tool\"><\/span>Installing ae.safetensors by Tool<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<h3>ComfyUI (Windows, macOS, Linux)<\/h3>\n<p>ComfyUI is the reference UI for FLUX. Place the file in the VAE folder:<\/p>\n<pre><code># Linux \/ macOS\nmv ae.safetensors ~\/ComfyUI\/models\/vae\/\n\n# Windows (PowerShell)\nMove-Item .ae.safetensors $HOMEComfyUImodelsvae<\/code><\/pre>\n<p>In the graph, add a <strong>Load VAE<\/strong> node and select <code>ae.safetensors<\/code> from the dropdown, then wire its <code>VAE<\/code> output into the <strong>VAE Decode<\/strong> node before the <strong>Save Image<\/strong> node. If you are using the all-in-one FLUX checkpoint (weights + VAE + text encoders in one file) from Comfy-Org, you do not need <code>ae.safetensors<\/code> separately \u2014 the VAE is baked in.<\/p>\n<h3>Forge \/ Automatic1111 \/ reForge<\/h3>\n<p>Put the file in <code>stable-diffusion-webui-forge\/models\/VAE\/<\/code> (Forge auto-detects it). In the Forge UI, set <em>UI<\/em> mode to <strong>flux<\/strong>, then in the <strong>VAE \/ Text Encoder<\/strong> selector at the top, pick <code>ae.safetensors<\/code> alongside the CLIP-L and T5-XXL encoders. Vanilla Automatic1111 does not officially support FLUX; use Forge instead.<\/p>\n<h3>SwarmUI \/ StableSwarmUI<\/h3>\n<p>Drop the file into <code>Models\/VAE\/<\/code>. Swarm auto-detects FLUX checkpoints and will select <code>ae.safetensors<\/code> automatically if it is the only FLUX-compatible VAE present.<\/p>\n<h3>diffusers (Python)<\/h3>\n<p>If you are loading the full FLUX repo from Hugging Face, the VAE is included and you do not touch <code>ae.safetensors<\/code> directly. To load a standalone <code>ae.safetensors<\/code>:<\/p>\n<pre><code>from diffusers import AutoencoderKL, FluxPipeline\nimport torch\n\nvae = AutoencoderKL.from_single_file(\n    \"ae.safetensors\", torch_dtype=torch.bfloat16\n)\npipe = FluxPipeline.from_pretrained(\n    \"black-forest-labs\/FLUX.1-schnell\",\n    vae=vae,\n    torch_dtype=torch.bfloat16,\n).to(\"cuda\")<\/code><\/pre>\n<p>Check the current <a href=\"https:\/\/huggingface.co\/docs\/diffusers\/main\/en\/api\/pipelines\/flux\" rel=\"noopener\" target=\"_blank\">diffusers FLUX pipeline docs<\/a> \u2014 the exact class name for the FLUX VAE has changed between diffusers releases, so pin your version.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"VRAM_and_Performance_Notes\"><\/span>VRAM and Performance Notes<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>The VAE itself is tiny (~335 MB in fp16, ~168 MB in fp8) \u2014 it is not what makes FLUX heavy. The transformer is the cost driver: FLUX.1-dev is ~23 GB in bf16 and ~12 GB in fp8. On a 24 GB card (RTX 3090\/4090) you can run FLUX with the VAE resident; on 12 GB cards you typically offload the VAE to CPU between decode calls.<\/p>\n<p>For a general sense of how much GPU memory diffusion and LLM checkpoints consume, use the <a href=\"https:\/\/convly.ai\/llm-vram-calculator\/\">Convly VRAM calculator<\/a> and the <a href=\"https:\/\/convly.ai\/best-gpus-for-local-llms-2026\/\">best GPUs for local models<\/a> guide.<\/p>\n<h3>Decoding at High Resolution<\/h3>\n<p>The FLUX VAE peaks in VRAM at decode time proportional to output resolution. A 2048\u00d72048 decode roughly quadruples the peak versus 1024\u00d71024. If you hit OOM only at the final decode step, enable tiled VAE:<\/p>\n<ul>\n<li><strong>ComfyUI:<\/strong> replace <em>VAE Decode<\/em> with <em>VAE Decode (Tiled)<\/em>.<\/li>\n<li><strong>diffusers:<\/strong> <code>pipe.vae.enable_tiling()<\/code> before inference.<\/li>\n<li><strong>Forge:<\/strong> tick <em>Tiled VAE<\/em> in the settings panel.<\/li>\n<\/ul>\n<h2><span class=\"ez-toc-section\" id=\"Common_aesafetensors_Errors\"><\/span>Common ae.safetensors Errors<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<table>\n<thead>\n<tr>\n<th>Error<\/th>\n<th>Cause<\/th>\n<th>Fix<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><code>Error while deserializing header: HeaderTooLarge<\/code><\/td>\n<td>Truncated download<\/td>\n<td>Re-download; verify SHA-256<\/td>\n<\/tr>\n<tr>\n<td>Black \/ green output image<\/td>\n<td>Wrong VAE (SDXL VAE loaded for FLUX)<\/td>\n<td>Confirm <code>ae.safetensors<\/code> is selected, not <code>sdxl_vae.safetensors<\/code><\/td>\n<\/tr>\n<tr>\n<td><code>Missing key(s) in state_dict<\/code><\/td>\n<td>Loading FLUX VAE into SD 1.5\/SDXL pipeline<\/td>\n<td>Use <code>FluxPipeline<\/code>\/<code>AutoencoderKL.from_single_file<\/code>, not <code>StableDiffusionPipeline<\/code><\/td>\n<\/tr>\n<tr>\n<td>File is 335 MB but rejected as invalid<\/td>\n<td>Downloaded HTML gate page instead of file<\/td>\n<td>Accept the license on the model page or use <code>huggingface-cli<\/code> with a token<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h2><span class=\"ez-toc-section\" id=\"aesafetensors_vs_the_All-in-One_FLUX_Checkpoint\"><\/span>ae.safetensors vs the All-in-One FLUX Checkpoint<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>You will see two distribution styles on Hugging Face and Civitai:<\/p>\n<ul>\n<li><strong>Split files<\/strong> (official BFL layout): <code>flux1-dev.safetensors<\/code> (transformer) + <code>ae.safetensors<\/code> (VAE) + <code>clip_l.safetensors<\/code> + <code>t5xxl_fp16.safetensors<\/code>. Preferred for diffusers, ComfyUI advanced graphs and finetuning \u2014 you can swap the transformer without re-downloading 20 GB of text encoders.<\/li>\n<li><strong>All-in-one<\/strong> (Comfy-Org repackage, e.g. <code>flux1-dev-fp8.safetensors<\/code>): every component bundled into one file. Simpler for one-click ComfyUI workflows; you do not need <code>ae.safetensors<\/code> as a separate file.<\/li>\n<\/ul>\n<h2><span class=\"ez-toc-section\" id=\"Where_FLUX_Fits_in_the_Broader_Model_Landscape\"><\/span>Where FLUX Fits in the Broader Model Landscape<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>FLUX is a local, self-hosted image model. If you are choosing between running it on your own GPU versus calling a hosted image or video API \u2014 Sora 2, Veo 3.1, Kling 2.5, Wan 2.5 \u2014 the trade-off is fixed hardware cost vs per-second billing. See the <a href=\"https:\/\/convly.ai\/self-hosting-vs-api-calculator\/\">self-hosting vs API break-even calculator<\/a> to estimate the crossover, and the <a href=\"https:\/\/convly.ai\/models\/\">Convly models database<\/a> for current API pricing across providers.<\/p>\n<p>For text models, similar VAE-style file splits do not apply \u2014 LLMs ship transformer weights only. If you are also running local LLMs alongside FLUX, tools like <a href=\"https:\/\/convly.ai\/what-is-ollama-complete-guide-2026\/\">Ollama<\/a> and <a href=\"https:\/\/convly.ai\/lm-studio-complete-guide-2026\/\">LM Studio<\/a> handle model management the way ComfyUI handles diffusion assets.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"Frequently_Asked_Questions\"><\/span>Frequently Asked Questions<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<h3>Is ae.safetensors the same file for FLUX.1-dev and FLUX.1-schnell?<\/h3>\n<p>Yes \u2014 the VAE architecture and weights are identical between FLUX.1-dev and FLUX.1-schnell. The two repos ship the same <code>ae.safetensors<\/code>, only under different licenses (schnell is Apache-2.0, dev is non-commercial). You can use either copy interchangeably.<\/p>\n<h3>Can I use the SDXL VAE with FLUX to save VRAM?<\/h3>\n<p>No. FLUX uses a 16-channel latent space; SDXL uses 4 channels. Loading the SDXL VAE against FLUX latents produces solid-color or noise output, and diffusers will usually raise a shape mismatch error. Use <code>ae.safetensors<\/code>.<\/p>\n<h3>Is ae.safetensors safe to download from third parties?<\/h3>\n<p>The safetensors format itself cannot execute code on load, so a corrupt or malicious file cannot pwn your machine the way a pickled <code>.ckpt<\/code> could. That said, a tampered VAE could produce degraded images. Prefer the <a href=\"https:\/\/huggingface.co\/black-forest-labs\" rel=\"noopener\" target=\"_blank\">official Black Forest Labs repos<\/a> and verify the SHA-256 hash.<\/p>\n<h3>Do I need ae.safetensors if I only use FLUX via an API?<\/h3>\n<p>No. Hosted APIs (Replicate, fal.ai, BFL&#8217;s own API, Together) run the full pipeline server-side. You only need <code>ae.safetensors<\/code> when running FLUX locally in ComfyUI, Forge, SwarmUI or diffusers.<\/p>\n<h3>Why is it named &ldquo;ae&rdquo; and not &ldquo;vae&rdquo;?<\/h3>\n<p>Black Forest Labs&#8217; internal naming uses &ldquo;ae&rdquo; (autoencoder) rather than the more common &ldquo;vae&rdquo; label from the Stable Diffusion ecosystem. Functionally it plays the same role: encoding pixels to latents and decoding latents back to pixels. The FLUX autoencoder is technically closer to a plain KL-regularized AE than the SD 1.5 VAE, but every UI treats it as a VAE.<\/p>\n<h3>Can I finetune ae.safetensors?<\/h3>\n<p>You can, but it is almost never worth it. The VAE is trained on billions of images and finetuning it typically degrades the transformer&#8217;s alignment with the latent space, hurting quality. Community FLUX finetunes and LoRAs modify only the transformer and reuse the stock <code>ae.safetensors<\/code>.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>ae.safetensors is the VAE (variational autoencoder) weight file shipped with Black Forest Labs&#8217; FLUX.1 image models. It decodes FLUX&#8217;s latents [\u2026]<\/p>\n","protected":false},"author":1,"featured_media":2741,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[9],"tags":[],"class_list":["post-2740","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-tutorials"],"_links":{"self":[{"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/posts\/2740","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/comments?post=2740"}],"version-history":[{"count":1,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/posts\/2740\/revisions"}],"predecessor-version":[{"id":2742,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/posts\/2740\/revisions\/2742"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/media\/2741"}],"wp:attachment":[{"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/media?parent=2740"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/categories?post=2740"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/tags?post=2740"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}