From 0644544d7407b989579609755c8a4867c1147402 Mon Sep 17 00:00:00 2001
From: Sun Haoran <haoransun@microsoft.com>
Date: Wed, 5 Nov 2025 19:00:28 +0000
Subject: [PATCH 1/2] init result visual

---
 .gitignore                  | 11 +++++++++++
 docs/_config.yml            |  4 ++++
 docs/_data/leaderboard.json | 32 +++++++++++++++++++++++++++++++
 docs/index.md               | 38 +++++++++++++++++++++++++++++++++++++
 4 files changed, 85 insertions(+)
 create mode 100644 docs/_config.yml
 create mode 100644 docs/_data/leaderboard.json
 create mode 100644 docs/index.md

diff --git a/.gitignore b/.gitignore
index d30146382..c31612e41 100644
--- a/.gitignore
+++ b/.gitignore
@@ -266,3 +266,14 @@ dmypy.json
 # Evaluation Result files
 evaluation_results/
 verification_results/
+
+# Jekyll (GitHub Pages) - docs folder
+docs/_site/
+docs/.sass-cache/
+docs/.jekyll-cache/
+docs/.jekyll-metadata
+docs/vendor/
+docs/.bundle/
+docs/tmp
+docs/Gemfile
+docs/Gemfile.lock
diff --git a/docs/_config.yml b/docs/_config.yml
new file mode 100644
index 000000000..e3f7e03d4
--- /dev/null
+++ b/docs/_config.yml
@@ -0,0 +1,4 @@
+title: BC-Bench
+description: Inspired by SWE-Bench, for Business Central (AL) ecosystem.
+show_downloads: false
+theme: jekyll-theme-cayman
diff --git a/docs/_data/leaderboard.json b/docs/_data/leaderboard.json
new file mode 100644
index 000000000..6c681ede7
--- /dev/null
+++ b/docs/_data/leaderboard.json
@@ -0,0 +1,32 @@
+[
+  {
+    "agent_name": "GitHub Copilot CLI",
+    "model": "gpt-4o",
+    "total": 19,
+    "resolved": 5,
+    "average_duration": 200.5,
+    "average_prompt_tokens": 15000.0,
+    "average_completion_tokens": 2500.0,
+    "date": "2025-11-05"
+  },
+  {
+    "agent_name": "Mini BC Agent",
+    "model": "gpt-4o-mini",
+    "total": 19,
+    "resolved": 3,
+    "average_duration": 150.2,
+    "average_prompt_tokens": 12000.0,
+    "average_completion_tokens": 2000.0,
+    "date": "2025-11-04"
+  },
+  {
+    "agent_name": "GitHub Copilot VS Code",
+    "model": "gpt-4o",
+    "total": 19,
+    "resolved": 7,
+    "average_duration": 250.8,
+    "average_prompt_tokens": 18000.0,
+    "average_completion_tokens": 3000.0,
+    "date": "2025-11-03"
+  }
+]
diff --git a/docs/index.md b/docs/index.md
new file mode 100644
index 000000000..12aea34f2
--- /dev/null
+++ b/docs/index.md
@@ -0,0 +1,38 @@
+---
+layout: default
+title: BC-Bench
+---
+
+## Evaluation Results
+
+The table below shows the performance of different AI coding agents on BC-Bench tasks. Results are sorted by accuracy (number of resolved tasks).
+
+<table>
+  <thead>
+    <tr>
+      <th>Agent (Model)</th>
+      <th>Accuracy</th>
+      <th>Avg Duration (s)</th>
+      <th>Token Consumption</th>
+      <th>Date</th>
+    </tr>
+  </thead>
+  <tbody>
+    {% assign sorted_results = site.data.leaderboard | sort: "resolved" | reverse %}
+    {% for result in sorted_results %}
+    <tr>
+      <td><strong>{{ result.agent_name }}</strong> ({{ result.model }})</td>
+      <td>{{ result.resolved }} / {{ result.total }} ({{ result.resolved | times: 100.0 | divided_by: result.total | round: 1 }}%)</td>
+      <td>{{ result.average_duration | round: 1 }}</td>
+      <td>{{ result.average_prompt_tokens | round: 0 }} prompt + {{ result.average_completion_tokens | round: 0 }} completion</td>
+      <td>{{ result.date }}</td>
+    </tr>
+    {% endfor %}
+  </tbody>
+</table>
+
+## About BC-Bench
+
+TODO
+
+For more information, visit the [BC-Bench repository](https://github.com/microsoft/BC-Bench).

From 5a6d6a78111f0d967096f03abdcb7ad6e9b755a1 Mon Sep 17 00:00:00 2001
From: Sun Haoran <haoransun@microsoft.com>
Date: Wed, 5 Nov 2025 19:28:52 +0000
Subject: [PATCH 2/2] manual update from previous runs

---
 docs/_data/leaderboard.json | 56 +++++++++++++++++++++++++------------
 1 file changed, 38 insertions(+), 18 deletions(-)

diff --git a/docs/_data/leaderboard.json b/docs/_data/leaderboard.json
index 6c681ede7..ad51016aa 100644
--- a/docs/_data/leaderboard.json
+++ b/docs/_data/leaderboard.json
@@ -1,32 +1,52 @@
 [
   {
     "agent_name": "GitHub Copilot CLI",
-    "model": "gpt-4o",
+    "model": "claude-sonnet-4",
     "total": 19,
-    "resolved": 5,
-    "average_duration": 200.5,
-    "average_prompt_tokens": 15000.0,
-    "average_completion_tokens": 2500.0,
+    "resolved": 7,
+    "average_duration": 0,
+    "average_prompt_tokens": 0,
+    "average_completion_tokens": 0,
     "date": "2025-11-05"
   },
   {
-    "agent_name": "Mini BC Agent",
-    "model": "gpt-4o-mini",
+    "agent_name": "GitHub Copilot CLI",
+    "model": "claude-sonnet-4.5",
     "total": 19,
-    "resolved": 3,
-    "average_duration": 150.2,
-    "average_prompt_tokens": 12000.0,
-    "average_completion_tokens": 2000.0,
+    "resolved": 9,
+    "average_duration": 0,
+    "average_prompt_tokens": 0,
+    "average_completion_tokens": 0,
     "date": "2025-11-04"
   },
   {
-    "agent_name": "GitHub Copilot VS Code",
-    "model": "gpt-4o",
+    "agent_name": "GitHub Copilot CLI",
+    "model": "claude-haiku-4.5",
     "total": 19,
-    "resolved": 7,
-    "average_duration": 250.8,
-    "average_prompt_tokens": 18000.0,
-    "average_completion_tokens": 3000.0,
-    "date": "2025-11-03"
+    "resolved": 10,
+    "average_duration": 0,
+    "average_prompt_tokens": 0,
+    "average_completion_tokens": 0,
+    "date": "2025-11-04"
+  },
+  {
+    "agent_name": "GitHub Copilot CLI",
+    "model": "gpt-5",
+    "total": 19,
+    "resolved": 5,
+    "average_duration": 0,
+    "average_prompt_tokens": 0,
+    "average_completion_tokens": 0,
+    "date": "2025-11-04"
+  },
+  {
+    "agent_name": "mini-bc-agent",
+    "model": "azure/gpt-4.1",
+    "total": 19,
+    "resolved": 6,
+    "average_duration": 0,
+    "average_prompt_tokens": 0,
+    "average_completion_tokens": 0,
+    "date": "2025-11-05"
   }
 ]