Hugging Face's logo

Buckets:
hf-doc-build
/
doc-dev

Files
xet
hf-doc-build/doc-dev / sagemaker /pr_2709 /en /agents /agent-harnesses.html
HuggingFaceDocBuilder's picture
HuggingFaceDocBuilder
download
raw
59.4 kB
<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Use AWS-hosted models with agent harnesses&quot;,&quot;local&quot;:&quot;use-aws-hosted-models-with-agent-harnesses&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Before you begin&quot;,&quot;local&quot;:&quot;before-you-begin&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Connect to SageMaker AI or JumpStart&quot;,&quot;local&quot;:&quot;connect-to-sagemaker-ai-or-jumpstart&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Pi&quot;,&quot;local&quot;:&quot;pi&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Hermes Agent&quot;,&quot;local&quot;:&quot;hermes-agent&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Tau&quot;,&quot;local&quot;:&quot;tau&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Connect to Amazon Bedrock&quot;,&quot;local&quot;:&quot;connect-to-amazon-bedrock&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Pi&quot;,&quot;local&quot;:&quot;pi&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Hermes Agent&quot;,&quot;local&quot;:&quot;hermes-agent&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Tau&quot;,&quot;local&quot;:&quot;tau&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Use another agent harness&quot;,&quot;local&quot;:&quot;use-another-agent-harness&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Troubleshooting&quot;,&quot;local&quot;:&quot;troubleshooting&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Examples&quot;,&quot;local&quot;:&quot;examples&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/>
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/entry/start.D0qXhpFD.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/chunks/D1ekGocw.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/chunks/BAlBQgxm.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/entry/app.BJw2fAxO.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/chunks/CShA6w_M.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/chunks/Ds9Xs-bq.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/chunks/DsnmJJEf.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/chunks/BdEMyugS.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/nodes/0.BLk_T2Pd.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/chunks/BchC4-7E.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/nodes/2.BscU0pCv.js" rel="modulepreload">
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/chunks/NrhFkieU.js" rel="modulepreload">
 <!--1vpulk2--><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Use AWS-hosted models with agent harnesses&quot;,&quot;local&quot;:&quot;use-aws-hosted-models-with-agent-harnesses&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Before you begin&quot;,&quot;local&quot;:&quot;before-you-begin&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Connect to SageMaker AI or JumpStart&quot;,&quot;local&quot;:&quot;connect-to-sagemaker-ai-or-jumpstart&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Pi&quot;,&quot;local&quot;:&quot;pi&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Hermes Agent&quot;,&quot;local&quot;:&quot;hermes-agent&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Tau&quot;,&quot;local&quot;:&quot;tau&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Connect to Amazon Bedrock&quot;,&quot;local&quot;:&quot;connect-to-amazon-bedrock&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Pi&quot;,&quot;local&quot;:&quot;pi&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Hermes Agent&quot;,&quot;local&quot;:&quot;hermes-agent&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Tau&quot;,&quot;local&quot;:&quot;tau&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Use another agent harness&quot;,&quot;local&quot;:&quot;use-another-agent-harness&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Troubleshooting&quot;,&quot;local&quot;:&quot;troubleshooting&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Examples&quot;,&quot;local&quot;:&quot;examples&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/><!---->
 <link href="/docs/sagemaker/pr_2709/en/_app/immutable/assets/0.tn0RQdqM.css" rel="modulepreload"> <!--[--><!--[0--><!--[--><!--[0--><!--[--><p></p> <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"><div class="inline-flex rounded-md max-sm:rounded-sm"><button class="inline-flex items-center gap-1 h-7 max-sm:h-7 px-2 max-sm:px-1.5 text-sm font-medium text-gray-800 border border-r-0 rounded-l-md max-sm:rounded-l-sm border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-live="polite"><span class="inline-flex items-center justify-center rounded-md p-0.5 max-sm:p-0 hover:text-gray-800 dark:hover:text-gray-200"><svg class="sm:size-3.5 size-3" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----></span> <span>Copy page</span></button> <button class="inline-flex items-center justify-center w-6 max-sm:w-5 h-7 max-sm:h-7 disabled:pointer-events-none text-sm text-gray-500 hover:text-gray-700 dark:hover:text-white rounded-r-md max-sm:rounded-r-sm border border-l transition border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-haspopup="menu" aria-expanded="false" aria-label="Open copy menu"><svg class="transition-transform text-gray-400 overflow-visible sm:size-3.5 size-3 rotate-0" width="1em" height="1em" viewBox="0 0 12 7" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M1 1L6 6L11 1" stroke="currentColor"></path></svg><!----></button></div> <!--[-1--><!--]--></div><!----> <!--[0--><h1 class="relative group"><a id="use-aws-hosted-models-with-agent-harnesses" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#use-aws-hosted-models-with-agent-harnesses"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Use AWS-hosted models with agent harnesses</span></h1><!--]--><!----> <p>Agent harnesses such as <a href="https://pi.dev/" rel="nofollow">Pi</a>, <a href="https://hermes-agent.nousresearch.com/" rel="nofollow">Hermes Agent</a>, or <a href="https://github.com/huggingface/tau" rel="nofollow">Tau</a> can use models hosted on Amazon SageMaker AI, SageMaker JumpStart, or Amazon Bedrock. The harness runs locally and sends model requests to an OpenAI-compatible endpoint on AWS.</p> <p>This guide covers client-side agent harnesses. <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/agents.html" rel="nofollow">Amazon Bedrock Agents</a> is a separate managed AWS service for building and orchestrating agents.</p> <p>The endpoint protocol and the model capability are separate requirements:</p> <ul><li>The endpoint must support streaming OpenAI Chat Completions.</li> <li>The model must reliably support tool calling. A model that can chat but cannot produce valid tool calls will not complete an agent loop.</li></ul> <!--[1--><h2 class="relative group"><a id="before-you-begin" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#before-you-begin"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Before you begin</span></h2><!--]--><!----> <p>For SageMaker AI and JumpStart, use a real-time endpoint backed by a container that implements <code>/v1/chat/completions</code>, such as the Hugging Face <a href="../get-started/dlcs">vLLM or SGLang DLC</a>. The IAM identity invoking the endpoint needs:</p> <ul><li><code>sagemaker:InvokeEndpoint</code> on the endpoint ARN.</li> <li><code>sagemaker:CallWithBearerToken</code> on <code>"*"</code>.</li></ul> <p>For Amazon Bedrock, select a model that supports the <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/apis.html" rel="nofollow">Chat Completions API</a> and create an <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/api-keys.html" rel="nofollow">Amazon Bedrock API key</a>. Prefer short-term keys for production; long-term keys are intended for exploration.</p> <!--[1--><h2 class="relative group"><a id="connect-to-sagemaker-ai-or-jumpstart" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#connect-to-sagemaker-ai-or-jumpstart"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Connect to SageMaker AI or JumpStart</span></h2><!--]--><!----> <p>SageMaker AI and JumpStart deployments use the same runtime URL. JumpStart deploys the selected model as a SageMaker AI endpoint, so the connection settings depend on the resulting endpoint, not how it was created.</p> <p>The base URL for a single-model endpoint is:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-text "><!---->https://runtime.sagemaker.&lt;REGION&gt;.amazonaws.com/endpoints/&lt;ENDPOINT_NAME&gt;/openai/v1<!----></pre></div><!----> <p>SageMaker AI authenticates OpenAI-compatible clients with short-lived bearer tokens. Generate one with the SageMaker Python SDK:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-python "><!----><span class="hljs-keyword">from</span> datetime <span class="hljs-keyword">import</span> timedelta
<span class="hljs-keyword">from</span> sagemaker.core.token_generator <span class="hljs-keyword">import</span> generate_token
token = generate_token(
 region=<span class="hljs-string">&quot;us-west-2&quot;</span>,
 expiry=timedelta(hours=<span class="hljs-number">1</span>),
)<!----></pre></div><!----> <p>Tokens are valid for up to 12 hours and cannot outlive the AWS credentials used to create them. Generate them at the point of use instead of saving them in a configuration file.</p> <!--[2--><h3 class="relative group"><a id="pi" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#pi"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Pi</span></h3><!--]--><!----> <p>Pi can resolve an API key by running a shell command. Save this script as <code>~/.pi/agent/sagemaker-token.py</code> so Pi can request a fresh token when it connects:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-python "><!----><span class="hljs-keyword">from</span> sagemaker.core.token_generator <span class="hljs-keyword">import</span> generate_token
<span class="hljs-built_in">print</span>(generate_token(region=<span class="hljs-string">&quot;us-west-2&quot;</span>))<!----></pre></div><!----> <p>Add the provider to <code>~/.pi/agent/models.json</code>:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-json "><!----><span class="hljs-punctuation">{</span>
 <span class="hljs-attr">&quot;providers&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-punctuation">{</span>
 <span class="hljs-attr">&quot;sagemaker&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-punctuation">{</span>
 <span class="hljs-attr">&quot;baseUrl&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;https://runtime.sagemaker.us-west-2.amazonaws.com/endpoints/my-endpoint/openai/v1&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;api&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;openai-completions&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;apiKey&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;!python ~/.pi/agent/sagemaker-token.py&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;models&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-punctuation">[</span>
 <span class="hljs-punctuation">{</span>
 <span class="hljs-attr">&quot;id&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;Qwen/Qwen3.8-27B&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;name&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;Qwen3.8 27B on SageMaker&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;reasoning&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-literal"><span class="hljs-keyword">false</span></span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;input&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-punctuation">[</span><span class="hljs-string">&quot;text&quot;</span><span class="hljs-punctuation">]</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;contextWindow&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-number">32768</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;maxTokens&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-number">4096</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;compat&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-punctuation">{</span>
 <span class="hljs-attr">&quot;supportsDeveloperRole&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-literal"><span class="hljs-keyword">false</span></span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;supportsReasoningEffort&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-literal"><span class="hljs-keyword">false</span></span>
 <span class="hljs-punctuation">}</span>
 <span class="hljs-punctuation">}</span>
 <span class="hljs-punctuation">]</span>
 <span class="hljs-punctuation">}</span>
 <span class="hljs-punctuation">}</span>
<span class="hljs-punctuation">}</span><!----></pre></div><!----> <p>Replace the Region, endpoint name, model ID, and model limits with your deployment values. Open Pi’s model picker and select the SageMaker model. See <a href="https://pi.dev/docs/latest/models" rel="nofollow">Pi custom models</a> for the complete provider schema.</p> <!--[2--><h3 class="relative group"><a id="hermes-agent" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#hermes-agent"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Hermes Agent</span></h3><!--]--><!----> <p>Hermes supports OpenAI-compatible endpoints through its <code>openai-api</code> provider. Export a fresh token in the shell that starts Hermes:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-bash "><!----><span class="hljs-built_in">export</span> OPENAI_BASE_URL=<span class="hljs-string">&quot;https://runtime.sagemaker.us-west-2.amazonaws.com/endpoints/my-endpoint/openai/v1&quot;</span>
<span class="hljs-built_in">export</span> OPENAI_API_KEY=<span class="hljs-string">&quot;<span class="hljs-subst">$(python -c &#x27;from sagemaker.core.token_generator import generate_token; print(generate_token(region=\&quot;us-west-2\&quot;)</span>)&#x27;)&quot;</span><!----></pre></div><!----> <p>Then configure the model in <code>~/.hermes/config.yaml</code>:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-yaml "><!----><span class="hljs-attr">model:</span>
 <span class="hljs-attr">default:</span> <span class="hljs-string">Qwen/Qwen3.8-27B</span>
 <span class="hljs-attr">provider:</span> <span class="hljs-string">openai-api</span><!----></pre></div><!----> <p>The environment variable keeps the short-lived token out of the configuration file. Generate a new token when it expires. See <a href="https://hermes-agent.nousresearch.com/docs/integrations/providers" rel="nofollow">Hermes AI providers</a> for other configuration methods.</p> <!--[2--><h3 class="relative group"><a id="tau" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#tau"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Tau</span></h3><!--]--><!----> <p>Tau supports OpenAI-compatible endpoints through custom providers. Because SageMaker bearer tokens are short-lived, pass the token through an environment variable generated in the shell that starts Tau:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-bash "><!----><span class="hljs-built_in">export</span> SAGEMAKER_API_KEY=<span class="hljs-string">&quot;<span class="hljs-subst">$(python -c &#x27;from sagemaker.core.token_generator import generate_token; print(generate_token(region=\&quot;us-west-2\&quot;)</span>)&#x27;)&quot;</span><!----></pre></div><!----> <p>Add the provider to <code>~/.tau/catalog.toml</code>:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-toml "><!----><span class="hljs-attr">schema_version</span> = <span class="hljs-number">1</span>
<span class="hljs-section">[[providers]]</span>
<span class="hljs-attr">name</span> = <span class="hljs-string">&quot;sagemaker&quot;</span>
<span class="hljs-attr">display_name</span> = <span class="hljs-string">&quot;Amazon SageMaker&quot;</span>
<span class="hljs-attr">kind</span> = <span class="hljs-string">&quot;openai-compatible&quot;</span>
<span class="hljs-attr">base_url</span> = <span class="hljs-string">&quot;https://runtime.sagemaker.us-west-2.amazonaws.com/endpoints/my-endpoint/openai/v1&quot;</span>
<span class="hljs-attr">api_key_env</span> = <span class="hljs-string">&quot;SAGEMAKER_API_KEY&quot;</span>
<span class="hljs-attr">credential_name</span> = <span class="hljs-string">&quot;sagemaker&quot;</span>
<span class="hljs-attr">models</span> = [<span class="hljs-string">&quot;Qwen/Qwen3.8-27B&quot;</span>]
<span class="hljs-attr">default_model</span> = <span class="hljs-string">&quot;Qwen/Qwen3.8-27B&quot;</span>
<span class="hljs-section">[providers.context_windows]</span>
<span class="hljs-attr">&quot;Qwen/Qwen3.8-27B&quot;</span> = <span class="hljs-number">32768</span><!----></pre></div><!----> <p>Replace the Region, endpoint name, model ID, and context window with your deployment values, then start Tau with the provider:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-bash "><!---->tau --provider sagemaker<!----></pre></div><!----> <p>The environment variable keeps the short-lived token out of the configuration file. Generate a new token when it expires. You can also run <code>/login custom</code> inside Tau for an interactive setup. See <a href="https://twotimespi.dev/guides/providers-and-models/" rel="nofollow">Tau providers and models</a> for the complete provider schema.</p> <!--[1--><h2 class="relative group"><a id="connect-to-amazon-bedrock" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#connect-to-amazon-bedrock"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Connect to Amazon Bedrock</span></h2><!--]--><!----> <p>Amazon Bedrock exposes an OpenAI-compatible endpoint at:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-text "><!---->https://bedrock-runtime.&lt;REGION&gt;.amazonaws.com/openai/v1<!----></pre></div><!----> <p>Set your Bedrock API key as an environment variable:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-bash "><!----><span class="hljs-built_in">export</span> AWS_BEARER_TOKEN_BEDROCK=<span class="hljs-string">&quot;&lt;BEDROCK_API_KEY&gt;&quot;</span><!----></pre></div><!----> <!--[2--><h3 class="relative group"><a id="pi" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#pi"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Pi</span></h3><!--]--><!----> <p>Add a Bedrock provider to <code>~/.pi/agent/models.json</code>:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-json "><!----><span class="hljs-punctuation">{</span>
 <span class="hljs-attr">&quot;providers&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-punctuation">{</span>
 <span class="hljs-attr">&quot;bedrock&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-punctuation">{</span>
 <span class="hljs-attr">&quot;baseUrl&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;https://bedrock-runtime.us-east-1.amazonaws.com/openai/v1&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;api&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;openai-completions&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;apiKey&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;$AWS_BEARER_TOKEN_BEDROCK&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;models&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-punctuation">[</span>
 <span class="hljs-punctuation">{</span>
 <span class="hljs-attr">&quot;id&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;openai.gpt-oss-120b&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;name&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-string">&quot;GPT OSS 120B on Bedrock&quot;</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;reasoning&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-literal"><span class="hljs-keyword">true</span></span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;input&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-punctuation">[</span><span class="hljs-string">&quot;text&quot;</span><span class="hljs-punctuation">]</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;contextWindow&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-number">131072</span><span class="hljs-punctuation">,</span>
 <span class="hljs-attr">&quot;maxTokens&quot;</span><span class="hljs-punctuation">:</span> <span class="hljs-number">8192</span>
 <span class="hljs-punctuation">}</span>
 <span class="hljs-punctuation">]</span>
 <span class="hljs-punctuation">}</span>
 <span class="hljs-punctuation">}</span>
<span class="hljs-punctuation">}</span><!----></pre></div><!----> <p>Use a model ID and limits supported by Bedrock in your Region.</p> <!--[2--><h3 class="relative group"><a id="hermes-agent" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#hermes-agent"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Hermes Agent</span></h3><!--]--><!----> <p>Point the OpenAI-compatible provider at Bedrock:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-bash "><!----><span class="hljs-built_in">export</span> OPENAI_BASE_URL=<span class="hljs-string">&quot;https://bedrock-runtime.us-east-1.amazonaws.com/openai/v1&quot;</span>
<span class="hljs-built_in">export</span> OPENAI_API_KEY=<span class="hljs-string">&quot;<span class="hljs-variable">$AWS_BEARER_TOKEN_BEDROCK</span>&quot;</span><!----></pre></div><!----> <p>Configure the selected model in <code>~/.hermes/config.yaml</code>:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-yaml "><!----><span class="hljs-attr">model:</span>
 <span class="hljs-attr">default:</span> <span class="hljs-string">openai.gpt-oss-120b</span>
 <span class="hljs-attr">provider:</span> <span class="hljs-string">openai-api</span><!----></pre></div><!----> <!--[2--><h3 class="relative group"><a id="tau" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#tau"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Tau</span></h3><!--]--><!----> <p>Add a Bedrock provider to <code>~/.tau/catalog.toml</code>:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-toml "><!----><span class="hljs-attr">schema_version</span> = <span class="hljs-number">1</span>
<span class="hljs-section">[[providers]]</span>
<span class="hljs-attr">name</span> = <span class="hljs-string">&quot;bedrock&quot;</span>
<span class="hljs-attr">display_name</span> = <span class="hljs-string">&quot;Amazon Bedrock&quot;</span>
<span class="hljs-attr">kind</span> = <span class="hljs-string">&quot;openai-compatible&quot;</span>
<span class="hljs-attr">base_url</span> = <span class="hljs-string">&quot;https://bedrock-runtime.us-east-1.amazonaws.com/openai/v1&quot;</span>
<span class="hljs-attr">api_key_env</span> = <span class="hljs-string">&quot;AWS_BEARER_TOKEN_BEDROCK&quot;</span>
<span class="hljs-attr">credential_name</span> = <span class="hljs-string">&quot;bedrock&quot;</span>
<span class="hljs-attr">models</span> = [<span class="hljs-string">&quot;openai.gpt-oss-120b&quot;</span>]
<span class="hljs-attr">default_model</span> = <span class="hljs-string">&quot;openai.gpt-oss-120b&quot;</span>
<span class="hljs-section">[providers.context_windows]</span>
<span class="hljs-attr">&quot;openai.gpt-oss-120b&quot;</span> = <span class="hljs-number">131072</span><!----></pre></div><!----> <p>Then start Tau with the provider:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-bash "><!---->tau --provider bedrock<!----></pre></div><!----> <p>Use a model ID and limits supported by Bedrock in your Region.</p> <!--[1--><h2 class="relative group"><a id="use-another-agent-harness" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#use-another-agent-harness"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Use another agent harness</span></h2><!--]--><!----> <p>For any harness that supports OpenAI Chat Completions, provide the same three values:</p> <ol><li>The SageMaker AI or Bedrock base URL.</li> <li>A SageMaker bearer token or Bedrock API key.</li> <li>The deployed model ID.</li></ol> <p>Confirm that the client supports streaming responses and tool calls. OpenAI-compatible text generation alone is not sufficient for an agent workflow.</p> <!--[1--><h2 class="relative group"><a id="troubleshooting" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#troubleshooting"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Troubleshooting</span></h2><!--]--><!----> <ul><li><strong>401 or 403 response</strong>: refresh an expired token, verify that the token and endpoint use the same Region, and check the invocation permissions.</li> <li><strong>404 response</strong>: use the base URL ending in <code>/openai/v1</code>; do not append <code>/chat/completions</code> when the harness adds it automatically.</li> <li><strong>The model answers but never calls tools</strong>: verify that the model supports tool calling and that the serving container has tool calling enabled. For vLLM, configure the appropriate tool-call parser for the model.</li> <li><strong>Malformed tool calls</strong>: reduce the tool schema complexity, verify the model’s expected chat template, and test a single tool before running a larger agent workflow.</li> <li><strong>Streaming errors</strong>: confirm that the container returns Chat Completions streams as server-sent events.</li></ul> <!--[1--><h2 class="relative group"><a id="examples" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#examples"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Examples</span></h2><!--]--><!----> <ul><li><a href="../examples/sagemaker-sdk-trip-planner-agent-vllm">Build a reasoning trip-planning agent with vLLM</a> demonstrates tool calling and MCP with Strands Agents on a SageMaker endpoint.</li> <li><a href="../examples/sagemaker-sdk-neuron-agent-inf">Deploy an AI agent on AWS Inferentia2</a> runs an agent with a model served on Inferentia2.</li> <li><a href="../examples/sagemaker-sdk-grpo-llm-trl">Improve tool calling with GRPO</a> fine-tunes a model to produce more reliable tool calls.</li></ul> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/hub-docs/blob/main/docs/sagemaker/source/agents/agent-harnesses.md" target="_blank"><svg class="mr-1" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M31,16l-7,7l-1.41-1.41L28.17,16l-5.58-5.59L24,9l7,7z"></path><path d="M1,16l7-7l1.41,1.41L3.83,16l5.58,5.59L8,23l-7-7z"></path><path d="M12.419,25.484L17.639,6.552l1.932,0.518L14.351,26.002z"></path></svg><!----> <span><span class="underline">Update</span> on GitHub</span></a><!----> <p></p><!--]--><!----><!--]--><!--]--><!--]--> <!--[-1--><!--]--><!--]-->
<script>
 {
 __sveltekit_18n89o7 = {
 base: "/docs/sagemaker/pr_2709/en",
 assets: "/docs/sagemaker/pr_2709/en"
 };
 const element = document.currentScript.parentElement;
 Promise.all([
 import("/docs/sagemaker/pr_2709/en/_app/immutable/entry/start.D0qXhpFD.js"),
 import("/docs/sagemaker/pr_2709/en/_app/immutable/entry/app.BJw2fAxO.js")
 ]).then(([kit, app]) => {
 kit.start(app, element, {
 node_ids: [0, 2],
 data: [null,null],
 form: null,
 error: null
 });
 });
 }
 </script>

Xet Storage Details

Size:
59.4 kB
·
Xet hash:
513477be19a2372fac4836bfab20860016414d714fc9d061aa531481f14e4a12

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.