Spaces:

openenvarena
/

arena

Running on CPU Upgrade

Launch video: posttrain.com on the closing card

#15

by xdotli - opened 24 days ago

←

Files changed (2) hide show

index.html CHANGED Viewed

@@ -285,8 +285,8 @@
     </div>
     <p>OpenEnvArena is a challenge to build better reinforcement learning environments, run by <a href="https://www.benchflow.ai/" target="_blank" rel="noopener">BenchFlow</a> on <a href="https://posttrain.com" target="_blank" rel="noopener">PostTrain Arena</a>. It ranks environments by how much they help a model learn. You and your coding agent build the tasks, interactions and rewards; the arena measures how well that learning transfers to new problems.</p>
     <p>An environment should include tasks with clear goals and starting states, the tools or actions the model can use, the observations it receives, and a reward function that checks whether it succeeded. It should reset each task to a fresh state, return feedback after each action, and end each episode with a reward between 0 and 1. Package the working OpenEnv server and its dependencies as a public Docker image, and share the tasks, supporting data and reward description as a Hugging Face dataset.</p>
-    <video class="about-video" id="launchVideo" controls playsinline preload="metadata" data-src="/open-env-arena-launch.mp4?v=921bc8d8a0b0" aria-label="OpenEnvArena launch video">
-      Your browser does not support embedded video. <a href="/open-env-arena-launch.mp4?v=921bc8d8a0b0">Watch the launch video</a>.
     </video>
     <h4 class="about-heading">How the challenge works</h4>
     <ol class="steps">

     </div>
     <p>OpenEnvArena is a challenge to build better reinforcement learning environments, run by <a href="https://www.benchflow.ai/" target="_blank" rel="noopener">BenchFlow</a> on <a href="https://posttrain.com" target="_blank" rel="noopener">PostTrain Arena</a>. It ranks environments by how much they help a model learn. You and your coding agent build the tasks, interactions and rewards; the arena measures how well that learning transfers to new problems.</p>
     <p>An environment should include tasks with clear goals and starting states, the tools or actions the model can use, the observations it receives, and a reward function that checks whether it succeeded. It should reset each task to a fresh state, return feedback after each action, and end each episode with a reward between 0 and 1. Package the working OpenEnv server and its dependencies as a public Docker image, and share the tasks, supporting data and reward description as a Hugging Face dataset.</p>
+    <video class="about-video" id="launchVideo" controls playsinline preload="metadata" data-src="/open-env-arena-launch.mp4?v=82081669080c" aria-label="OpenEnvArena launch video">
+      Your browser does not support embedded video. <a href="/open-env-arena-launch.mp4?v=82081669080c">Watch the launch video</a>.
     </video>
     <h4 class="about-heading">How the challenge works</h4>
     <ol class="steps">

open-env-arena-launch.mp4 CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:921bc8d8a0b07fd5ec2b54f996cdf7637b9ef72ff2f2b130e977b4b90a363e35
-size 43131782

 version https://git-lfs.github.com/spec/v1
+oid sha256:82081669080c66147de72b3e74701d9a0536b2ed5d5504d4aba8244b295880e3
+size 21180807