From 180cf5bc86c33f25d4226cdfc29f39e4c854ac53 Mon Sep 17 00:00:00 2001
From: daniel5u <175603571+daniel5u@users.noreply.github.com>
Date: Wed, 25 Mar 2026 01:18:13 +0000
Subject: [PATCH] update data from SWE Benchmark

---
 src/content/benchmarks/swe_bash_only.json | 138 +++++-----------------
 1 file changed, 29 insertions(+), 109 deletions(-)

diff --git a/src/content/benchmarks/swe_bash_only.json b/src/content/benchmarks/swe_bash_only.json
index fd6a0f8..039c50c 100644
--- a/src/content/benchmarks/swe_bash_only.json
+++ b/src/content/benchmarks/swe_bash_only.json
@@ -8,7 +8,7 @@
         "Coding",
         "Agent"
     ],
-    "lastUpdated": "2026-01-12",
+    "lastUpdated": "2026-03-25",
     "metrics": {
         "unit": "% Resolved",
         "isBetterHigher": true
@@ -19,140 +19,60 @@
     },
     "snapshot": [
         {
-            "modelRef": "anthropic/claude-opus-4-5-thinking",
-            "score": 74.4
+            "modelRef": "anthropic/claude-4-5-opus-high-reasoning",
+            "score": 76.8
         },
         {
-            "modelRef": "google/gemini-3-pro",
-            "score": 74.2
-        },
-        {
-            "modelRef": "openai/gpt-5-2",
-            "score": 71.8
-        },
-        {
-            "modelRef": "anthropic/claude-4-5-sonnet",
-            "score": 70.6
-        },
-        {
-            "modelRef": "openai/gpt-5-2",
-            "score": 69.0
-        },
-        {
-            "modelRef": "anthropic/claude-4-opus",
-            "score": 67.6
-        },
-        {
-            "modelRef": "openai/gpt-5-1-codex",
-            "score": 66.0
-        },
-        {
-            "modelRef": "openai/gpt-5-1",
-            "score": 66.0
-        },
-        {
-            "modelRef": "openai/gpt-5-medium",
-            "score": 65.0
-        },
-        {
-            "modelRef": "anthropic/claude-4-sonnet",
-            "score": 64.93
-        },
-        {
-            "modelRef": "kimi/kimi-k2-thinking",
-            "score": 63.4
-        },
-        {
-            "modelRef": "minimax/minimax-m2",
-            "score": 61.0
+            "modelRef": "google/gemini-3-flash-high-reasoning",
+            "score": 75.8
         },
         {
-            "modelRef": "deepseek/deepseek-v3-2-reasoning",
-            "score": 60.0
+            "modelRef": "minimax/minimax-m2-5-high-reasoning",
+            "score": 75.8
         },
         {
-            "modelRef": "openai/gpt-5-mini-medium",
-            "score": 59.8
+            "modelRef": "anthropic/claude-opus-4-6",
+            "score": 75.6
         },
         {
-            "modelRef": "openai/o3",
-            "score": 58.4
+            "modelRef": "openai/gpt-5-2-codex",
+            "score": 72.8
         },
         {
-            "modelRef": "mistral/devstral-small",
-            "score": 56.4
+            "modelRef": "z-ai/glm-5-high-reasoning",
+            "score": 72.8
         },
         {
-            "modelRef": "alibaba/qwen3-coder-480b-a35b-instruct",
-            "score": 55.4
-        },
-        {
-            "modelRef": "zai/glm-4-6",
-            "score": 55.4
-        },
-        {
-            "modelRef": "zai/glm-4-5",
-            "score": 54.2
-        },
-        {
-            "modelRef": "mistral/devstral-2",
-            "score": 53.8
-        },
-        {
-            "modelRef": "google/gemini-2-5-pro",
-            "score": 53.6
-        },
-        {
-            "modelRef": "anthropic/claude-3-7-sonnet",
-            "score": 52.8
-        },
-        {
-            "modelRef": "openai/o4-mini",
-            "score": 45.0
-        },
-        {
-            "modelRef": "kimi/kimi-k2-0905",
-            "score": 43.8
-        },
-        {
-            "modelRef": "openai/gpt-4-1",
-            "score": 39.58
-        },
-        {
-            "modelRef": "openai/gpt-5-nano-medium",
-            "score": 34.8
-        },
-        {
-            "modelRef": "google/gemini-2-5-flash",
-            "score": 28.73
+            "modelRef": "openai/gpt-5-2",
+            "score": 72.8
         },
         {
-            "modelRef": "openai/gpt-oss-120b",
-            "score": 26.0
+            "modelRef": "openai/gpt-5-2-codex",
+            "score": 72.8
         },
         {
-            "modelRef": "openai/gpt-4-1-mini",
-            "score": 23.94
+            "modelRef": "anthropic/claude-4-5-sonnet-high-reasoning",
+            "score": 71.4
         },
         {
-            "modelRef": "openai/gpt-4o",
-            "score": 21.62
+            "modelRef": "kimi/kimi-k2-5-high-reasoning",
+            "score": 70.8
         },
         {
-            "modelRef": "meta/llama-4-maverick",
-            "score": 21.04
+            "modelRef": "deepseek/deepseek-v3-2-high-reasoning",
+            "score": 70.0
         },
         {
-            "modelRef": "google/gemini-2-0-flash",
-            "score": 13.52
+            "modelRef": "google/gemini-3-pro",
+            "score": 69.6
         },
         {
-            "modelRef": "meta/llama-4-scout",
-            "score": 9.06
+            "modelRef": "anthropic/claude-4-5-haiku-high-reasoning",
+            "score": 66.6
         },
         {
-            "modelRef": "alibaba/qwen2-5-coder-32b-instruct",
-            "score": 9.0
+            "modelRef": "openai/gpt-5-mini",
+            "score": 56.2
         }
     ]
 }
\ No newline at end of file