Inference
Explore
Leaderboard
Benchmarks
Labs
Run locally
GPUs
Search
⌘K
EN
English
Русский
Локально
中文
本地运行
21 Nov
2026
Benchmarks
/
cursorBench41
cursorBench41
In index
Coding
Best in coding
Models
?
Models with a result.
12
Top result
?
Best result on this benchmark.
65.4
%
Gemini 4 Neon
Top-3 spread
?
Points from first to third.
4.0
pts
Year
—
Result and price
0
20
40
60
80
100
$0.3
$1
$3
$10
$30
$100
RESULT
PARETO
OUTPUT PRICE PER 1M TOKENS · LOG SCALE
12 of 12 results
Download
PNG
Results
12 results
Export
CSV
Export
JSON
#
Model
Result
?
The score from the source.
01
Gemini 4 Neon
65.4
%
02
Gemini 4 Carbon
61.5
%
03
Gemini 4 Hydrogen
61.4
%
04
Claude Fable 5.2
61.3
%
05
Claude Opus 5.5
58.1
%
06
Gemini 4 Argon
56.5
%
07
Claude Sonnet 5.5
54.4
%
08
Grok 4.7
48.6
%
09
Claude Haiku 5.5
48.4
%
10
Grok 4.6
44.3
%
11
Muse Spark 1.3
41.9
%
12
GPT-5.6 Sol
40.9
%