{"id":787,"date":"2026-06-06T01:59:10","date_gmt":"2026-06-06T01:59:10","guid":{"rendered":"https:\/\/convly.ai\/best-local-llm-for-coding-2026\/"},"modified":"2026-06-10T05:04:43","modified_gmt":"2026-06-10T05:04:43","slug":"best-local-llm-for-coding-2026","status":"publish","type":"post","link":"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/","title":{"rendered":"The Best Local LLM for Coding in 2026 (Tested on Real Tasks)"},"content":{"rendered":"<p>Running a coding model locally means your proprietary code never touches someone else&#8217;s server \u2014 and you pay nothing per token. The catch has always been quality. In 2026, local coding models finally crossed the line from &#8220;toy&#8221; to &#8220;genuinely useful,&#8221; and this guide ranks the best of them by performance, hardware needs, and real-world coding behavior.<\/p>\n<p>To run any of these, you&#8217;ll want Ollama \u2014 see <a href=\"https:\/\/convly.ai\/pt\/what-is-ollama-complete-guide-2026\/\">what it is<\/a> e <a href=\"https:\/\/convly.ai\/pt\/how-to-install-ollama-2026\/\">how to install it<\/a>.<\/p>\n<div class=\"convly-tldr\">\n<h3>Principais conclus\u00f5es<\/h3>\n<ul>\n<li><strong>Best overall local coder:<\/strong> <strong>Qwen 3.6 27B<\/strong> \u2014 the strongest dense coding model at ~<strong>77.2% SWE-bench<\/strong>, needs ~22 GB VRAM.<\/li>\n<li><strong>Best for lighter hardware:<\/strong> <strong>Gemma 4 26B A4B<\/strong> or a smaller Qwen coder variant \u2014 solid code with a smaller footprint.<\/li>\n<li><strong>Frontier (if you can host it):<\/strong> <strong>Kimi K2.6<\/strong> \u2014 ~58.6 on SWE-Bench Pro, ties top cloud models, but needs heavy quantization for consumer hardware.<\/li>\n<li><strong>The honest truth:<\/strong> a top local coder rivals mid-tier cloud assistants; the very best cloud models still lead on the hardest, multi-file tasks.<\/li>\n<li><strong>Why bother:<\/strong> privacy, zero per-token cost, and offline work.<\/li>\n<\/ul>\n<\/div>\n<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_84 counter-flat ez-toc-counter ez-toc-container-direction\">\n<label for=\"ez-toc-cssicon-toggle-item-6a38ab803354e\" class=\"ez-toc-cssicon-toggle-label\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Alternar<\/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-6a38ab803354e\"  aria-label=\"Alternar\" \/><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\/best-local-llm-for-coding-2026\/#What_%E2%80%9Cbest%E2%80%9D_means_for_a_coding_model\" >What &#8220;best&#8221; means for a coding model<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#Best_overall_Qwen_36_27B\" >Best overall: Qwen 3.6 27B<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#Best_for_lighter_hardware_Gemma_4_26B_A4B\" >Best for lighter hardware: Gemma 4 26B A4B<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#Frontier_option_Kimi_K26\" >Frontier option: Kimi K2.6<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#How_they_compare\" >How they compare<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#Local_vs_cloud_coding_assistants_the_honest_take\" >Local vs cloud coding assistants: the honest take<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#Hooking_it_into_your_editor\" >Hooking it into your editor<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-8\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#Quantization_and_context_the_settings_that_make_or_break_the_result\" >Quantization and context: the settings that make or break the result<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-9\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#FAQ\" >Perguntas frequentes<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-10\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#Bottom_line\" >Conclus\u00e3o<\/a><\/li><li class='ez-toc-page-1'><a class=\"ez-toc-link ez-toc-heading-11\" href=\"https:\/\/convly.ai\/pt\/best-local-llm-for-coding-2026\/#Related_articles\" >Artigos relacionados<\/a><\/li><\/ul><\/nav><\/div>\n<h2><span class=\"ez-toc-section\" id=\"What_%E2%80%9Cbest%E2%80%9D_means_for_a_coding_model\"><\/span>What &#8220;best&#8221; means for a coding model<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Coding is a harsh test for an LLM because the output either runs or it doesn&#8217;t. The benchmark that matters most is <strong>SWE-bench<\/strong>, which measures whether a model can resolve real GitHub issues \u2014 not just autocomplete a line, but understand a codebase and ship a working fix. We weight three things:<\/p>\n<ol>\n<li><strong>SWE-bench performance<\/strong> \u2014 can it actually solve real engineering tasks?<\/li>\n<li><strong>Hardware fit<\/strong> \u2014 a brilliant model you can&#8217;t load is no help.<\/li>\n<li><strong>Behavior on real work<\/strong> \u2014 does it follow instructions, respect your style, and avoid hallucinating APIs?<\/li>\n<\/ol>\n<h2><span class=\"ez-toc-section\" id=\"Best_overall_Qwen_36_27B\"><\/span>Best overall: Qwen 3.6 27B<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p><strong>Qwen 3.6 27B<\/strong> is the local coding champion of 2026. As the strongest <em>dense<\/em> coding model available to self-host, it reaches roughly <strong>77.2% on SWE-bench<\/strong> and needs about <strong>22 GB of VRAM<\/strong> \u2014 meaning a 24 GB card (an RTX 4090, RTX 5090, or 7900 XTX) or Apple Silicon with enough unified memory can run it. In practice it handles multi-step refactors, writes coherent functions across files, and follows instructions tightly. It&#8217;s also Apache 2.0, so you can build commercial tools on it.<\/p>\n<pre><code>ollama run qwen3-coder\n<\/code><\/pre>\n<p>If you have the VRAM, this is the one to run.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"Best_for_lighter_hardware_Gemma_4_26B_A4B\"><\/span>Best for lighter hardware: Gemma 4 26B A4B<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Not everyone has 22 GB of VRAM. <strong>Gemma 4 26B A4B<\/strong> is a mixture-of-experts model that delivers strong coding help with a much friendlier memory footprint, plus built-in tool calling \u2014 handy for agentic coding workflows. For local coding without a high-end GPU, it&#8217;s the most practical starting point, and a smaller Qwen coder variant is a good fallback on tighter machines.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"Frontier_option_Kimi_K26\"><\/span>Frontier option: Kimi K2.6<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>If you have serious hardware and want the closest-to-cloud experience, <strong>Kimi K2.6<\/strong> reaches about <strong>58.6 on SWE-Bench Pro<\/strong> \u2014 a tougher benchmark than standard SWE-bench \u2014 effectively tying the top cloud models on hard engineering tasks. The cost is size: it needs heavy quantization to fit consumer hardware, and even then it&#8217;s demanding. For most people it&#8217;s overkill, but it shows how far open coding models have come.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"How_they_compare\"><\/span>How they compare<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<table class=\"convly-vs\">\n<thead>\n<tr>\n<th>Modelo<\/th>\n<th>Coding strength<\/th>\n<th>Hardware<\/th>\n<th>Melhor para<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Qwen 3.6 27B<\/td>\n<td>~77% SWE-bench<\/td>\n<td>~22 GB VRAM<\/td>\n<td>The best local coder most people can run<\/td>\n<\/tr>\n<tr>\n<td>Gemma 4 26B A4B<\/td>\n<td>Fortes<\/td>\n<td>Mid-range<\/td>\n<td>Lighter hardware, agentic workflows<\/td>\n<\/tr>\n<tr>\n<td>Kimi K2.6<\/td>\n<td>~58.6 SWE-Bench Pro<\/td>\n<td>Very high (quantized)<\/td>\n<td>Frontier quality, heavy rigs<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h2><span class=\"ez-toc-section\" id=\"Local_vs_cloud_coding_assistants_the_honest_take\"><\/span>Local vs cloud coding assistants: the honest take<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Should you ditch your cloud coding assistant? For most professionals, not entirely \u2014 yet. A top local model like Qwen 3.6 now rivals mid-tier cloud assistants and is genuinely productive for everyday coding, but the very best cloud models still pull ahead on the hardest, large-context, multi-file problems. The local case is strongest when <strong>privacy is non-negotiable<\/strong> (proprietary or regulated code), when you want <strong>zero per-token cost<\/strong> for high-volume use, or when you need to <strong>work offline<\/strong>. Many developers run both: local for sensitive or routine work, cloud for the gnarliest tasks. If you&#8217;re weighing the cloud side too, see our roundup of the <a href=\"https:\/\/convly.ai\/pt\/best-ai-coding-assistants\/\">melhores assistentes de programa\u00e7\u00e3o com IA<\/a>.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"Hooking_it_into_your_editor\"><\/span>Hooking it into your editor<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Once the model is running in Ollama, you can wire it into your workflow. Ollama&#8217;s <code>ollama launch<\/code> command sets up coding tools like Claude Code, OpenCode, and Codex against a local model with no config files, and most popular editor extensions accept a local OpenAI-compatible endpoint \u2014 point them at <code>http:\/\/localhost:11434<\/code> and you have an in-editor assistant that never sends your code to the cloud.<\/p>\n<p><!--ai-enriched--><\/p>\n<h2><span class=\"ez-toc-section\" id=\"Quantization_and_context_the_settings_that_make_or_break_the_result\"><\/span>Quantization and context: the settings that make or break the result<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>The model you pick matters less than how you run it. Two settings \u2014 the quantization level and the context window \u2014 quietly decide whether a local coding model feels like a capable pair-programmer or a frustrating autocomplete that invents functions. Most people who conclude &#8220;local models can&#8217;t code&#8221; simply ran a too-aggressive quant in a too-small context.<\/p>\n<p><strong>Quantization<\/strong> shrinks a model&#8217;s weights so it fits in your VRAM, and it trades a little accuracy for a lot of memory. For coding, the practical floor is <strong>Q4_K_M<\/strong>. At Q4, quality loss is modest and the memory savings are large \u2014 for most setups it is the sweet spot. Step up to Q5, Q6, or Q8 and you reclaim a few more percent of accuracy, but the returns shrink fast and the file roughly doubles by Q8. The real cliff is below Q4: at Q3 and Q2 a coding model starts emitting subtle syntax errors, mismatched brackets, and logic that looks right but isn&#8217;t \u2014 the worst failure mode, because it still compiles. The honest rule:<\/p>\n<ul>\n<li><strong>Q8 \/ Q6:<\/strong> best fidelity, for when you have VRAM to spare and want the model&#8217;s full ability \u2014 code and arithmetic-heavy logic hold up best here.<\/li>\n<li><strong>Q4_K_M:<\/strong> the default. Run this before you blame the model.<\/li>\n<li><strong>Below Q4:<\/strong> avoid for code. You are better off dropping to a smaller model at Q4 than a bigger one at Q2.<\/li>\n<\/ul>\n<p><strong>Janela de contexto<\/strong> is the other half. Coding agents have to hold your files, errors, and edit history in memory, and a long context is what lets a model reason across a whole module instead of a single snippet. The catch is that context is not free: the KV cache grows roughly linearly with length, so a generous window can eat several gigabytes \u2014 on large models, a 128K-token context can consume tens of gigabytes on its own. That memory competes directly with your model weights.<\/p>\n<p>So size the window to your hardware rather than maxing it out. As a rough guide, an 8&nbsp;GB card is comfortable around 4\u20138K tokens, 16&nbsp;GB stretches to 16\u201332K, and 24&nbsp;GB makes 64K+ practical. Setting a 128K window &#8220;just in case&#8221; usually backfires \u2014 it starves the weights, slows generation, and rarely helps day-to-day editing. If you need more headroom, enable <strong>KV-cache quantization<\/strong> (8-bit), which can roughly halve cache memory with little quality cost, and lean on tools like Aider&#8217;s repository map that compress a codebase into a small, high-signal summary instead of stuffing every file into the prompt.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"FAQ\"><\/span>Perguntas frequentes<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<h3>What is the best local LLM for coding in 2026?<\/h3>\n<p>Qwen 3.6 27B \u2014 it&#8217;s the strongest dense coding model you can self-host, at roughly 77% SWE-bench, needing about 22 GB of VRAM. On lighter hardware, Gemma 4 26B A4B is the most practical alternative.<\/p>\n<h3>Can a local LLM replace GitHub Copilot or Claude?<\/h3>\n<p>For routine and privacy-sensitive coding, yes \u2014 Qwen 3.6 is genuinely productive and keeps your code local. For the hardest multi-file tasks, the best cloud models still lead. A common setup is to use local models for sensitive or high-volume work and a cloud assistant for the toughest problems.<\/p>\n<h3>What hardware do I need to run a local coding model?<\/h3>\n<p>Qwen 3.6 27B wants about 22 GB of VRAM \u2014 a 24 GB GPU or Apple Silicon with ample unified memory. For 8\u201316 GB machines, use Gemma 4 or a smaller Qwen coder variant. See our <a href=\"https:\/\/convly.ai\/pt\/ollama-system-requirements-2026\/\">system requirements guide<\/a> for specifics.<\/p>\n<h3>Is Qwen better than DeepSeek for coding?<\/h3>\n<p>For pure coding throughput on self-hostable hardware, Qwen 3.6 27B is the stronger dedicated coder. DeepSeek&#8217;s R1 shines at step-by-step reasoning and math; it&#8217;s excellent when a problem needs careful logic, but Qwen is the more focused coding model.<\/p>\n<h3>How do I use a local coding model in VS Code?<\/h3>\n<p>Run the model in Ollama, then point a compatible editor extension at Ollama&#8217;s OpenAI-compatible endpoint (<code>http:\/\/localhost:11434<\/code>). Ollama&#8217;s <code>ollama launch<\/code> can also configure tools like Claude Code and Codex against your local model automatically.<\/p>\n<h3>What quantization level should I use for a local coding model?<\/h3>\n<p>Use Q4_K_M as your baseline \u2014 it keeps almost all of a model&#8217;s coding ability while fitting comfortably in VRAM, and it is the level most benchmarks and recommendations assume. Move up to Q6 or Q8 if you have memory to spare and want maximum fidelity, which matters most for arithmetic-heavy or tightly logical code. Avoid going below Q4 (Q3 or Q2) for code: the savings are small and you start getting syntax errors and subtle logic bugs. A smaller model at Q4 almost always beats a larger one squeezed to Q2.<\/p>\n<h3>How much context window do I need for coding locally?<\/h3>\n<p>More than you think for whole-file work, but far less than the model&#8217;s maximum. The context window holds your open files, errors, and the agent&#8217;s edit history, but it consumes VRAM that grows with its length, so it competes with the model weights. For most local coding, 16\u201332K tokens is plenty; reserve very large windows for repository-scale tasks and only if you have the memory. If you run out of room, turn on 8-bit KV-cache quantization or use a tool with a repository map rather than maxing the window.<\/p>\n<h3>Can a local model do inline autocomplete like Copilot, not just chat?<\/h3>\n<p>Yes. The fast tab-completion you get from Copilot relies on fill-in-the-middle (FIM), where the model completes code using both the text before and after your cursor. Coding-specialized models such as the Qwen-Coder family are trained for FIM, and editor extensions like Continue can route completions to your local model for low-latency, fully offline autocomplete. Plain general-purpose chat models are weaker at this, so for an autocomplete-first workflow pick a model that explicitly supports FIM and a smaller quant that keeps latency low.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"Bottom_line\"><\/span>Conclus\u00e3o<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Local coding models grew up in 2026. If you can spare ~22 GB of VRAM, Qwen 3.6 27B is the best local coder available and a real alternative to a cloud assistant for most work. On lighter hardware, Gemma 4 gets you most of the way. The pitch is simple: your code stays yours, you pay nothing per token, and the quality is finally good enough to mean it.<\/p>\n<p><!--related-block--><\/p>\n<div class=\"convly-related\">\n<h2><span class=\"ez-toc-section\" id=\"Related_articles\"><\/span>Artigos relacionados<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul>\n<li><a href=\"https:\/\/convly.ai\/pt\/claude-5-new-ai-models-june-2026\/\">Existe um Claude 5? Claude Fable 5 e todos os principais modelos de IA de junho de 2026<\/a><\/li>\n<li><a href=\"https:\/\/convly.ai\/pt\/llm-hallucinations-complete-guide\/\">Alucina\u00e7\u00f5es de LLMs em 2026: Por que ocorrem e como evit\u00e1-las<\/a><\/li>\n<li><a href=\"https:\/\/convly.ai\/pt\/prompt-engineering-techniques\/\">Engenharia de prompts em 2026: 12 t\u00e9cnicas que realmente funcionam<\/a><\/li>\n<li><a href=\"https:\/\/convly.ai\/pt\/what-is-ollama-complete-guide-2026\/\">O que \u00e9 o Ollama? Guia completo para executar LLMs localmente em 2026<\/a><\/li>\n<\/ul>\n<\/div>","protected":false},"excerpt":{"rendered":"<p>A local coding model means your code never leaves your machine. Here are the best ones in 2026 \u2014 ranked by SWE-bench, hardware needs, and how they handle real refactors.<\/p>","protected":false},"author":1,"featured_media":793,"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":[3],"tags":[624,623,628,626,627,625],"class_list":["post-787","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-llms","tag-best-coding-llm","tag-best-local-llm-for-coding","tag-code-llm","tag-local-coding-model","tag-ollama-coding","tag-qwen-coder"],"_links":{"self":[{"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/posts\/787","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=787"}],"version-history":[{"count":2,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/posts\/787\/revisions"}],"predecessor-version":[{"id":955,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/posts\/787\/revisions\/955"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/media\/793"}],"wp:attachment":[{"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/media?parent=787"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/categories?post=787"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/convly.ai\/pt\/wp-json\/wp\/v2\/tags?post=787"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}