CursorBench

Real coding tasks from Cursor users, graded on correctness in the Cursor agent harness. Published by Cursor.

CategoryCoding
MetricAccuracy
SourceEpoch AI (evaluator)
LicenseCC BY 4.0
Benchmark pagecursor.com
StatusScored
Weight1.3 within Coding
FitDiscrimination 0.65, noise 0.12

Results

#ModelProviderResultDate
1 Gemini 4 Carbon High Google 60.4% 2026-11-11
2 Gemini 4 Carbon Medium Google 60.1% 2026-11-11
3 Claude Fable 5.2 Max Anthropic 60.0% 2026-11-11
4 Gemini 4 Carbon Low Google 58.6% 2026-11-11
5 Claude Opus 5.5 Max Anthropic 57.8% 2026-11-11
6 Claude Fable 5.2 High Anthropic 56.9% 2026-11-11
7 Claude Fable 5.2 Extra High Anthropic 56.3% 2026-11-11
8 Claude Opus 5.5 High Anthropic 56.0% 2026-11-11
9 Claude Opus 5.5 Extra High Anthropic 56.0% 2026-11-11
10 Claude Sonnet 5.5 Max Anthropic 55.5% 2026-11-11
11 Claude Fable 5.2 Medium Anthropic 54.8% 2026-11-11
12 Claude Fable 5.2 Low Anthropic 53.2% 2026-11-11
13 Claude Sonnet 5.5 Extra High Anthropic 53.1% 2026-11-11
14 Gemini 4 Argon High Google 52.9% 2026-11-11
15 Gemini 4 Argon Medium Google 52.8% 2026-11-11
16 Claude Opus 5.5 Medium Anthropic 52.5% 2026-11-11
17 Claude Fable 5.1 Max Anthropic 51.8% 2026-11-11
18 Claude Fable 5.1 Extra High Anthropic 51.6% 2026-11-11
19 Gemini 4 Argon Low Google 50.6% 2026-11-11
20 Muse Wonder Max Meta 50.4% 2026-11-11
21 Claude Fable 5.1 High Anthropic 49.2% 2026-11-11
22 Gemini 4 Argon Minimal Google 48.8% 2026-11-11
23 Claude Haiku 5.5 Max Anthropic 48.4% 2026-11-11
24 Muse Wonder Extra High Meta 48.2% 2026-11-11
25 Claude Sonnet 5.5 High Anthropic 47.8% 2026-11-11
26 DeepSeek V4.1 Flash 1017 Max DeepSeek 47.7% 2026-11-11
27 Claude Fable 5.1 Medium Anthropic 46.8% 2026-11-11
28 Claude Opus 5 Max Anthropic 46.6% 2026-11-11
29 Grok 4.7 Extra High xAI 46.3% 2026-11-11
30 DeepSeek V4.1 Flash 1017 Medium DeepSeek 46.3% 2026-11-11
31 Claude Opus 5 Extra High Anthropic 46.1% 2026-11-11
32 Muse Wonder High Meta 46.1% 2026-11-11
33 DeepSeek V4.1 Flash 1017 High DeepSeek 45.6% 2026-11-11
34 Muse Wonder Medium Meta 45.2% 2026-11-11
35 Claude Fable 5.1 Low Anthropic 45.1% 2026-11-11
36 DeepSeek V4.1 Flash 1017 Low DeepSeek 44.9% 2026-11-11
37 Claude Opus 5 High Anthropic 44.7% 2026-11-11
38 Claude Haiku 5.5 Extra High Anthropic 44.3% 2026-11-11
39 Grok 4.7 High xAI 43.9% 2026-11-11
40 Claude Opus 5.5 Low Anthropic 43.7% 2026-11-11
41 Claude Opus 5 Medium Anthropic 43.3% 2026-11-11
42 GLM-5.3 Max Zhipu AI 42.6% 2026-11-11
43 Claude Haiku 5.5 High Anthropic 42.3% 2026-11-11
44 GPT-5.6 Sol Max OpenAI 41.7% 2026-11-11
45 Muse Wonder Low Meta 41.7% 2026-11-11
46 Muse Spark 1.3 Max Meta 41.6% 2026-11-11
47 Grok 4.7 Medium xAI 41.6% 2026-11-11
48 Grok 4.6 Extra High xAI 41.4% 2026-11-11
49 GPT-5.6 Terra Max OpenAI 41.3% 2026-11-11
50 Claude Opus 5 Low Anthropic 40.7% 2026-11-11
51 Grok 4.6 High xAI 40.4% 2026-11-11
52 Gemini 3.8 Flash High Google 39.6% 2026-11-11
53 Claude Sonnet 5.5 Medium Anthropic 39.2% 2026-11-11
54 GLM-5.3 High Zhipu AI 38.0% 2026-11-11
55 GPT-5.6 Sol Extra High OpenAI 37.7% 2026-11-11
56 Muse Spark 1.3 Extra High Meta 37.5% 2026-11-11
57 Gemini 3.8 Flash Medium Google 37.3% 2026-11-11
58 Claude Haiku 5.5 Medium Anthropic 36.9% 2026-11-11
59 GLM-5.3-Flash Max Zhipu AI 36.8% 2026-11-11
60 Grok 4.6 Medium xAI 36.1% 2026-11-11
61 GPT-5.6 Luna Max OpenAI 35.9% 2026-11-11
62 Claude Sonnet 5.5 Low Anthropic 35.8% 2026-11-11
63 GPT-5.6 Sol High OpenAI 35.7% 2026-11-11
64 Muse Wonder Minimal Meta 34.2% 2026-11-11
65 Claude Sonnet 5 Max Anthropic 34.1% 2026-11-11
66 GPT-5.6 Terra Extra High OpenAI 33.6% 2026-11-11
67 Grok 4.6 Low xAI 33.4% 2026-11-11
68 Muse Spark 1.3 High Meta 33.4% 2026-11-11
69 GLM-5.3 Low Zhipu AI 33.3% 2026-11-11
70 Grok 4.7 Low xAI 33.1% 2026-11-11
71 GPT-5.6 Luna Extra High OpenAI 33.0% 2026-11-11
72 Muse Spark 1.3 Medium Meta 32.6% 2026-11-11
73 Claude Sonnet 5 Extra High Anthropic 32.0% 2026-11-11
74 GPT-5.6 Sol Medium OpenAI 31.1% 2026-11-11
75 GLM-5.3-Flash High Zhipu AI 31.1% 2026-11-11
76 Claude Haiku 5.5 Low Anthropic 30.9% 2026-11-11
77 Claude Sonnet 5 High Anthropic 30.8% 2026-11-11
78 GPT-5.6 Terra High OpenAI 30.7% 2026-11-11
79 GPT-5.6 Luna High OpenAI 29.4% 2026-11-11
80 Muse Spark 1.3 Low Meta 29.3% 2026-11-11
81 Claude Sonnet 5 Medium Anthropic 28.0% 2026-11-11
82 Composer 2.5 Cursor 27.7% 2026-11-11
83 GPT-5.6 Terra Medium OpenAI 27.6% 2026-11-11
84 GLM-5.3-Flash Low Zhipu AI 26.9% 2026-11-11
85 GPT-5.6 Terra Low OpenAI 25.2% 2026-11-11
86 GPT-5.6 Sol Low OpenAI 24.6% 2026-11-11
87 Muse Spark 1.3 Minimal Meta 24.3% 2026-11-11
88 Claude Sonnet 5 Low Anthropic 24.1% 2026-11-11
89 GPT-5.6 Luna Medium OpenAI 22.2% 2026-11-11
90 GPT-5.6 Luna Low OpenAI 16.0% 2026-11-11