Update results

roboflow · Feb 19, 2025 · 0d4a37e · 0d4a37e
1 parent bcb1d2f
commit 0d4a37e
Show file tree

Hide file tree

Showing 2 changed files with 111 additions and 5 deletions.
diff --git a/index.html b/index.html
@@ -40,7 +40,7 @@ <h1>How's GPT O1 Doing?</h1>
                     <p>You can contribute your own tests, too! See the <a href="https://github.com/roboflow/gpt-checkup?tab=readme-ov-file#-contribute">GitHub README</a> for contributing instructions.</p>
                 </div>
                 <div class="header_subtitle">
-                    <p>Tests are run every day at 1am PT. Last updated February 18, 2025.</p>
+                    <p>Tests are run every day at 1am PT. Last updated February 19, 2025.</p>
                     <p>Made with ❤️ by the team at <a href="https://roboflow.com">Roboflow</a>.</p>
                 </div>
                 <div class="header_cta">
@@ -58,12 +58,12 @@ <h1>How's GPT O1 Doing?</h1>
                         <div class="feature_header" style="min-height: auto">
                             <div class="feature_header_text" style="gap: var(--spacing-sizing-4)">
                                 <h2>Response Time</h2>
-                                <p style="font-size: 16px; color: var(--gray-700)">Today, the average response time to receive results from our tests was <b>3.8 seconds</b> per request.</p>
+                                <p style="font-size: 16px; color: var(--gray-700)">Today, the average response time to receive results from our tests was <b>3.82 seconds</b> per request.</p>
                                 <p class="subtitle">This number only accounts for requests made by this application.</p>
                             </div>
                             <div class="chart">
                                 <div class="chart_box chart_box_green">
-                                    <p>3.8 s</p>
+                                    <p>3.82 s</p>
                                 </div>
                             </div>
                         </div>
@@ -176,7 +176,7 @@ <h3><span class="explainer_icon far fa-comment-dots"></span>Prompt</h3>
                                         <h3><span class="explainer_icon far fa-image"></span>Image</h3>
                                         <img class="test_image" src="images/swift.png" alt="Image of the input into GPT-4" />
                                         <h3><span class="explainer_icon far fa-sparkles"></span>Result</h3>
-                                        <pre></pre>
+                                        <pre>I was</pre>
                                         <p class="subtitle" style="margin-top: 16px; text-align: center">Test submitted by <a href="https://roboflow.com" target="_blank">Roboflow</a></p>
                                     </div>
                                 </div>
@@ -765,7 +765,7 @@ <h2>Zero Shot Classification</h2>
                                         </div>
                                     </div>
                                     <p class="result_text">Of the last 7 tests, conducted daily, this test has passed <b>100%</b> of the time.</p>
-                                    <p class="request_price"><i class="far fa-coins"></i>Today's request cost $0.01</p>
+                                    <p class="request_price"><i class="far fa-coins"></i>Today's request cost $0.008</p>
                                 </div>
                                 <div class="explainer_dropdown">
                                     <button type="button" class="dropdown dropdown_learn active">Learn about this test</button>

diff --git a/results/2025-02-19.json b/results/2025-02-19.json
@@ -0,0 +1,106 @@
+{
+    "zero_shot_classification": {
+        "score": 1,
+        "success": true,
+        "price": 0.008199999999999999,
+        "pass_fail": "Pass",
+        "response_time": 8.38942813873291,
+        "result": "Toyota Camry"
+    },
+    "count_fruit": {
+        "score": 0,
+        "success": false,
+        "price": 0.01545,
+        "pass_fail": "Fail",
+        "response_time": 7.712848663330078,
+        "result": ""
+    },
+    "document_ocr": {
+        "score": 0,
+        "success": false,
+        "price": 0.013989999999999999,
+        "pass_fail": "Fail",
+        "response_time": 8.065871477127075,
+        "result": "I was"
+    },
+    "handwriting_ocr": {
+        "score": 0,
+        "success": false,
+        "price": 0.015529999999999999,
+        "pass_fail": "Fail",
+        "response_time": 14.523216962814331,
+        "result": ""
+    },
+    "extraction_ocr": {
+        "score": 0,
+        "success": false,
+        "price": 0.013649999999999999,
+        "pass_fail": "Fail",
+        "response_time": 6.437875032424927,
+        "result": "Failed to produce a valid JSON output: "
+    },
+    "math_ocr": {
+        "score": 0,
+        "success": false,
+        "price": 0.02113,
+        "pass_fail": "Fail",
+        "response_time": 8.603970766067505,
+        "result": "Failed to produce a valid JSON output: "
+    },
+    "object_detection": {
+        "score": 0,
+        "success": false,
+        "price": 0.01584,
+        "pass_fail": "Fail",
+        "response_time": 7.256832599639893,
+        "result": "Failed to produce a valid JSON output: "
+    },
+    "graph_understanding": {
+        "score": 0,
+        "success": false,
+        "price": 0.0157,
+        "pass_fail": "Fail",
+        "response_time": 9.157629489898682,
+        "result": "Failed to produce a valid JSON output: "
+    },
+    "color_recognition": {
+        "score": 0,
+        "success": false,
+        "price": 0.0157,
+        "pass_fail": "Fail",
+        "response_time": 7.645347833633423,
+        "result": "Failed to produce a valid JSON output: "
+    },
+    "annotation_qa": {
+        "score": 0,
+        "success": false,
+        "price": 0.02135,
+        "pass_fail": "Fail",
+        "response_time": 9.838661670684814,
+        "result": "Failed to produce a valid JSON output: "
+    },
+    "measurement": {
+        "score": 0,
+        "success": false,
+        "price": 0.01566,
+        "pass_fail": "Fail",
+        "response_time": 10.821077108383179,
+        "result": "Failed to produce a valid JSON output: "
+    },
+    "easy_captcha": {
+        "score": 0,
+        "success": false,
+        "price": 0.01281,
+        "pass_fail": "Fail",
+        "response_time": 6.027362823486328,
+        "result": ""
+    },
+    "easy_captcha_persuade": {
+        "score": 0,
+        "success": false,
+        "price": 0.013309999999999999,
+        "pass_fail": "Fail",
+        "response_time": 9.13540267944336,
+        "result": ""
+    }
+}