Update leaders and resources

docusaurus.config.ts

@@ -98,6 +98,10 @@ const config: Config = {
               label: 'Craft vs Cruft',
               href: 'https://www.youtube.com/channel/UC4nEbAo5xFsOZDk2v0RIGHA',
             },
+            {
+              label: 'More',
+              href: '/resources',
+            },
           ],
         },
         {
@@ -108,13 +112,14 @@ const config: Config = {
             //   href: 'https://stackoverflow.com/questions/tagged/docusaurus',
             // },
             {
-              label: 'nopilot.dev Discord',
+              label: 'Nopilot Discord',
               href: 'https://discord.gg/k3hzFm5ykA',
             },
             {
-              label: 'Resources',
-              href: '/resources',
+              label: 'Nopilot YouTube',
+              href: 'https://www.youtube.com/@nopilot-dev',
             },
+
             // {
             //   label: 'Twitter',
             //   href: 'https://twitter.com/docusaurus',

src/components/HomepageFeatures/index.tsx

@@ -128,11 +128,14 @@ export default function HomepageFeatures(): JSX.Element {
             <div className="padding-horiz--md">
               <Heading as="h2">Updates</Heading>
               <div>
-                <p className=""><a href="/blog/dissecting-devin">Blog: Dissecting Devin</a></p>
+                <ul>
+                  <li><a href="https://www.youtube.com/watch?v=Ko-R3MtTpWQ">Reading of SWE-agent paper</a></li>
+                  <li><a href="/blog/dissecting-devin">Blog: Dissecting Devin</a></li>
+                </ul>
+              </div>
+              <div>
+                <iframe width="560" height="315" src="https://www.youtube.com/embed/aKrjE7NKfw8" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
               </div>
-
-              <iframe width="560" height="315" src="https://www.youtube.com/embed/jhkY_BUDVcU" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
-
 
             </div>
           </div>

src/pages/leaderboards.mdx

@@ -20,6 +20,30 @@ ML researcher [theblackcat102](https://github.com/theblackcat102) [reports](http
 
 Paul Gauthier [points out](https://github.com/princeton-nlp/SWE-bench/issues/72) that some SWE-bench cases appear to be underspecified and effectively impossible to solve because the tests rely on implementation detail. It's unclear what the maximum possible score is.
 
+## Aider Leaderboards
+
+The coding agent Aider maintains a [leaderboard](https://aider.chat/docs/leaderboards) of model performance within its key subtasks.
+
+### Code Editing
+
+- openai/gpt-4o
+- claude-3-opus
+- gpt-4 (0613)
+- gpt-4-turbo (2024-04-09)
+- deepseek-chat v2 (Open Weight)
+- gpt-3.5-turbo
+- gemini-1.5-pro
+- claude-3-sonnet
+- deepseek-coder (Open Weight)
+
+### Code refactoring
+
+- claude-3-opus
+- openai/gpt-4o
+- gpt-4 (1106-preview)
+- gemini-1.5-pro
+- gpt-4-turbo (2024-04-09)
+
 ## LiveCodeBench
 
 [LiveCodeBench](https://livecodebench.github.io/leaderboard.html): "Holistic and Contamination Free Evaluation of Large Language Models for Code"
@@ -31,10 +55,15 @@ Tests the strength of models across different coding sub-tasks.
 * Test Output Prediction
 * Code Execution
 
-*Last checked: 2024-04-10*
-* Proprietary Leaders: GPT-4-Turbo-2024-04-09, Claude-3-Opus
-* Open Weight Leaders: [WizardCoder-33B-V1.1](https://huggingface.co/WizardLM/WizardCoder-33B-V1.1), [deepseek-coder-33b-instruct](https://huggingface.co/deepseek-ai/deepseek-coder-33b-instruct), [CodeLlama-34b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-34b-Instruct-hf)
+The below listing of standout models across subtasks is subjective.
 
+*Last checked: 2024-05-14*
+* Proprietary Leaders: GPT-4o, GPT-4-Turbo, Claude-3-Opus
+* Open Weight Leaders:
+  * [LLama3-70b-Ins](https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct)
+  * [WizardCoder-33B-V1.1](https://huggingface.co/WizardLM/WizardCoder-33B-V1.1)
+  * [deepseek-coder-33b-instruct](https://huggingface.co/deepseek-ai/deepseek-coder-33b-instruct)
+  * [Phind-34B-V2](https://huggingface.co/Phind/Phind-CodeLlama-34B-v2)
 
 ## Other notable benchmarks
 

src/pages/resources.md

@@ -6,25 +6,31 @@ title: Resources
 
 ## Community
 
-* [Can we beat Devin? Discord](https://discord.gg/canwebeatdevin): shared space with several teams
+* [nopilot.dev Discord](https://discord.gg/k3hzFm5ykA) - Discussion about the ecosystem
 * [OpenDevin Discord](https://discord.gg/mBuDGRzzES)
-* [nopilot.dev Discord](https://discord.gg/k3hzFm5ykA): discussion about this hub
+* [SWE-agent Discord](https://discord.gg/AVEFbBn2rH)
+* OpenDevAI Discord
 
-## Autonomous Coders (WebUX)
+## Videos
+* [nopilot.dev YouTube Channel](https://www.youtube.com/@nopilot-dev)
+* [Playlist on Autonomous DevTools](https://www.youtube.com/playlist?list=PLUBjHzmgsFNf_9LrJlk2t0n7pGiOLVqoX)
+
+## Coding Agents (WebUX)
 
-* Devin by Cognition
 * [OpenDevin](https://github.com/OpenDevin/OpenDevin)
+* Devin by Cognition
 * [Devika](https://github.com/stitionai/devika)
 * [Anterion](https://github.com/MiscellaneousStuff/anterion): UX wrapping SWE-agent
 
-## Autonomous Coders (Command-line)
+## Coding Agents (Backend)
 * [AutoCodeRover](https://github.com/nus-apr/auto-code-rover): from NUS-apr, highest score on SWE-bench lite
 * [SWE-agent](https://swe-agent.com) from Princeton NLP, first Open Source agent to break 10% SWE-bench
+* [Sweep](https://sweep.dev): Turn bugs into pull requests
 
 [Longer list](https://github.com/e2b-dev/awesome-ai-agents) by E2B.
 
 ## Eval Tools
 * [SWE-bench](https://www.swebench.com/)
 * [moatless-tools](https://github.com/aorwall/moatless-tools)
-* [SWE-bench-util](https://github.com/raymyers/swe-bench-util)
+