SShogo Bench

CursorBench

Real coding-agent sessions from Cursor, graded on whether the change does what the developer asked.

Agent Index weight
4%
in the Agentic coding pillar
Models tested
22
from the benchmark’s leaderboard and reports
Normalised for the index
0 → 0, 1 → 100
Official site

Results

#ModelScoreNormalisedHarnessCost/taskDate
1Gemini 4 Carbon Supported
Google
60.4%source$9.22—
2Claude Fable 5.2 Supported
Anthropic
60.0%source$32.79—
3Claude Opus 5.5 Supported
Anthropic
57.8%source$13.43—
4Gemini 4 Hydrogen Estimated
Google
57.8%source$5.86—
5Claude Sonnet 5.5 Estimated
Anthropic
55.5%source$7.05—
6Gemini 4 Argon Supported
Google
52.9%source$4.57—
7Claude Fable 5.1 Supported
Anthropic
51.8%source$17.28—
8Muse Wonder Estimated
Meta
50.4%source$5.25—
9Claude Haiku 5.5 Estimated
Anthropic
48.4%source$1.12—
10DeepSeek V4.1 Flash 1017 Estimated
DeepSeek
47.7%source$0.98—
11Claude Opus 5 Supported
Anthropic
46.6%source$11.95—
12Grok 4.7 Estimated
xAI
46.3%source$6.01—
13GLM-5.3 Supported
Z.ai (Zhipu)
42.6%source$5.05—
14GPT-5.6 Sol Supported
OpenAI
41.7%source$8.23—
15Muse Spark 1.3 Estimated
Meta
41.6%source$2.64—
16Grok 4.6 Supported
xAI
41.4%source$6.10—
17GPT-5.6 Terra Estimated
OpenAI
41.3%source$5.14—
18Gemini 3.8 Flash Supported
Google
39.6%source$4.70—
19GLM-5.3-Flash Supported
Z.ai (Zhipu)
36.8%source$0.39—
20GPT-5.6 Luna Estimated
OpenAI
35.9%source$1.03—
21Claude Sonnet 5 Supported
Anthropic
34.1%source$7.17—
22Composer 2.5
Cursor
27.7%source$0.68—

For models without a CursorBench result, the index estimates one from Epoch’s Capabilities Index using a logistic fit on 20 models (residual SD 0.16 on the logit scale). Estimates are labelled and widen the model’s range.