CursorBench 4.0

We evaluate agents on ambiguous, multi-file tasks from real Cursor sessions. Higher scores are better.

	Model
1	Gemini 4 Carbon High	60.4%	$9.22	39,960	71
2	Gemini 4 Carbon Medium	60.1%	$8.49	37,067	67
3	Fable 5.2 Max	60.0%	$32.79	352,794	218
4	Gemini 4 Carbon Low	58.6%	$11.89	42,663	124
5	Opus 5.5 Max	57.8%	$13.43	218,363	185
6	Fable 5.2 High	56.9%	$9.70	99,259	69
7	Fable 5.2 Extra High	56.3%	$15.48	161,251	108
8	Opus 5.5 Extra High	56.0%	$6.98	101,083	109
9	Opus 5.5 High	56.0%	$3.97	53,078	68
10	Sonnet 5.5 Max	55.5%	$7.05	271,920	170
11	Fable 5.2 Medium	54.8%	$4.66	44,103	35
12	Fable 5.2 Low	53.2%	$3.94	36,616	30
13	Sonnet 5.5 Extra High	53.1%	$2.81	100,158	78
14	Gemini 4 Argon High	52.9%	$4.57	109,500	118
15	Gemini 4 Argon Medium	52.8%	$4.39	103,600	115
16	Opus 5.5 Medium	52.5%	$2.91	37,954	54
17	Fable 5.1 Max	51.8%	$17.28	117,236	128
18	Fable 5.1 Extra High	51.6%	$13.01	87,294	101
19	Gemini 4 Argon Low	50.6%	$4.09	94,200	110
20	Muse Wonder Max	50.4%	$5.25	49,692	114
21	Fable 5.1 High	49.2%	$9.08	58,438	77
22	Haiku 5.5 Max	48.4%	$1.12	325,934	163
23	Muse Wonder Extra High	48.2%	$4.24	39,564	96
24	Sonnet 5.5 High	47.8%	$1.20	37,391	41
25	DeepSeek V4.1 Flash 1017 Max	47.7%	$0.98	537,047	187
26	Fable 5.1 Medium	46.8%	$7.05	45,411	63
27	Opus 5 Max	46.6%	$11.95	85,384	106
28	Grok 4.7 Extra High	46.3%	$6.01	70,141	88
29	DeepSeek V4.1 Flash 1017 Medium	46.3%	$0.80	398,255	189
30	Opus 5 Extra High	46.1%	$11.43	80,094	103
31	Muse Wonder High	46.1%	$3.51	33,612	84
32	DeepSeek V4.1 Flash 1017 High	45.6%	$0.77	385,196	186
33	Muse Wonder Medium	45.2%	$2.82	27,124	75
34	Fable 5.1 Low	45.1%	$5.44	34,795	51
35	DeepSeek V4.1 Flash 1017 Low	44.9%	$0.60	290,651	161
36	Opus 5 High	44.7%	$9.00	61,405	86
37	Haiku 5.5 Extra High	44.3%	$0.56	143,813	93
38	Grok 4.7 High	43.9%	$4.69	56,382	71
39	Opus 5.5 Low	43.7%	$1.17	15,811	28
40	Opus 5 Medium	43.3%	$6.94	45,272	72
41	GLM 5.3 Max	42.6%	$5.05	96,387	166
42	Haiku 5.5 High	42.3%	$0.32	77,057	59
43	GPT-5.6 Sol Max	41.7%	$8.23	42,944	99
44	Muse Wonder Low	41.7%	$1.91	18,880	56
45	Grok 4.7 Medium	41.6%	$3.49	36,683	60
46	Muse Spark 1.3 Max	41.6%	$2.64	52,005	98
47	Grok 4.6 Extra High	41.4%	$6.10	49,814	56
48	GPT-5.6 Terra Max	41.3%	$5.14	60,814	107
49	Opus 5 Low	40.7%	$4.87	31,995	57
50	Grok 4.6 High	40.4%	$5.20	41,387	48
51	Gemini 3.8 Flash High	39.6%	$4.70	162,565	324
52	Sonnet 5.5 Medium	39.2%	$0.52	16,036	22
53	GLM 5.3 High	38.0%	$3.24	60,031	114
54	GPT-5.6 Sol Extra High	37.7%	$4.40	24,729	55
55	Muse Spark 1.3 Extra High	37.5%	$2.10	40,891	83
56	Gemini 3.8 Flash Medium	37.3%	$4.06	128,364	290
57	Haiku 5.5 Medium	36.9%	$0.17	42,659	39
58	GLM 5.3 Flash Max	36.8%	$0.39	56,410	118
59	Grok 4.6 Medium	36.1%	$3.48	24,893	40
60	GPT-5.6 Luna Max	35.9%	$1.03	87,284	208
61	Sonnet 5.5 Low	35.8%	$0.37	11,668	18
62	GPT-5.6 Sol High	35.7%	$2.85	16,174	41
63	Muse Wonder Minimal	34.2%	$0.84	9,761	33
64	Sonnet 5 Max	34.1%	$7.17	149,257	140
65	GPT-5.6 Terra Extra High	33.6%	$1.81	23,436	43
66	Grok 4.6 Low	33.4%	$2.25	16,307	32
67	Muse Spark 1.3 High	33.4%	$1.66	30,654	69
68	GLM 5.3 Low	33.3%	$2.04	31,983	81
69	Grok 4.7 Low	33.1%	$1.58	15,677	40
70	GPT-5.6 Luna Extra High	33.0%	$0.44	40,598	98
71	Muse Spark 1.3 Medium	32.6%	$1.49	27,255	64
72	Sonnet 5 Extra High	32.0%	$4.55	83,373	102
73	GPT-5.6 Sol Medium	31.1%	$1.77	10,111	32
74	GLM 5.3 Flash High	31.1%	$0.25	35,104	84
75	Haiku 5.5 Low	30.9%	$0.08	23,382	25
76	Sonnet 5 High	30.8%	$3.48	61,146	85
77	GPT-5.6 Terra High	30.7%	$1.11	13,162	33
78	GPT-5.6 Luna High	29.4%	$0.25	23,368	64
79	Muse Spark 1.3 Low	29.3%	$0.93	17,483	47
80	Sonnet 5 Medium	28.0%	$2.31	39,114	65
81	Composer 2.5	27.7%	$0.68	17,347	41
82	GPT-5.6 Terra Medium	27.6%	$0.64	7,307	25
83	GLM 5.3 Flash Low	26.9%	$0.15	17,831	58
84	GPT-5.6 Terra Low	25.2%	$0.52	5,914	23
85	GPT-5.6 Sol Low	24.6%	$0.87	4,885	21
86	Muse Spark 1.3 Minimal	24.3%	$0.56	10,620	34
87	Sonnet 5 Low	24.1%	$1.39	23,772	46
88	GPT-5.6 Luna Medium	22.2%	$0.08	7,642	32
89	GPT-5.6 Luna Low	16.0%	$0.03	3,288	18

Changelog

Oct 7, 2026

Reporting

Updated Sonnet 5.5 results to account for adjusted pricing.

Sep 10, 2026

Tasks

CursorBench 4.0
- Introduced new long-horizon problems focused on edit, refactor, investigation, intent understanding, managing jobs, and design adherence.

Aug 11, 2026

Reporting

Updated Sonnet 5 results to account for adjusted pricing.

Jul 30, 2026

Reporting

Updated GPT-5.6 Terra and Luna results to account for adjusted pricing.

Jul 9, 2026

Reporting

Updated GPT-5.6 Sol, Terra, and Luna results to account for cache write costs.

Jul 8, 2026

Tasks

CursorBench 3.2
- Introduced instruction following and advanced tool use problems.

May 19, 2026

Tasks

CursorBench 3.1
- Introduced problems focused on codebase understanding, bugfinding, planning, and code review.
- Improved grading criteria for some edit tasks.

Mar 11, 2026

Tasks

CursorBench 3.0
- Initial set of tasks focused on edit, refactor, and bugfix problems.

Avg cost / task is computed by applying each model's published per-million-token pricing (input, cache read, cache write, and output) to the tokens it used on each task. Results are subject to variance; small differences in scores may not be statistically meaningful.

CursorBench 4.0

We evaluate agents on ambiguous, multi-file tasks from real Cursor sessions. Higher scores are better.

More about CursorBench

	Model
1	Gemini 4 Carbon High	60.4%	$9.22	39,960	71
2	Gemini 4 Carbon Medium	60.1%	$8.49	37,067	67
3	Fable 5.2 Max	60.0%	$32.79	352,794	218
4	Gemini 4 Carbon Low	58.6%	$11.89	42,663	124
5	Opus 5.5 Max	57.8%	$13.43	218,363	185
6	Fable 5.2 High	56.9%	$9.70	99,259	69
7	Fable 5.2 Extra High	56.3%	$15.48	161,251	108
8	Opus 5.5 Extra High	56.0%	$6.98	101,083	109
9	Opus 5.5 High	56.0%	$3.97	53,078	68
10	Sonnet 5.5 Max	55.5%	$7.05	271,920	170
11	Fable 5.2 Medium	54.8%	$4.66	44,103	35
12	Fable 5.2 Low	53.2%	$3.94	36,616	30
13	Sonnet 5.5 Extra High	53.1%	$2.81	100,158	78
14	Gemini 4 Argon High	52.9%	$4.57	109,500	118
15	Gemini 4 Argon Medium	52.8%	$4.39	103,600	115
16	Opus 5.5 Medium	52.5%	$2.91	37,954	54
17	Fable 5.1 Max	51.8%	$17.28	117,236	128
18	Fable 5.1 Extra High	51.6%	$13.01	87,294	101
19	Gemini 4 Argon Low	50.6%	$4.09	94,200	110
20	Muse Wonder Max	50.4%	$5.25	49,692	114
21	Fable 5.1 High	49.2%	$9.08	58,438	77
22	Haiku 5.5 Max	48.4%	$1.12	325,934	163
23	Muse Wonder Extra High	48.2%	$4.24	39,564	96
24	Sonnet 5.5 High	47.8%	$1.20	37,391	41
25	DeepSeek V4.1 Flash 1017 Max	47.7%	$0.98	537,047	187
26	Fable 5.1 Medium	46.8%	$7.05	45,411	63
27	Opus 5 Max	46.6%	$11.95	85,384	106
28	Grok 4.7 Extra High	46.3%	$6.01	70,141	88
29	DeepSeek V4.1 Flash 1017 Medium	46.3%	$0.80	398,255	189
30	Opus 5 Extra High	46.1%	$11.43	80,094	103
31	Muse Wonder High	46.1%	$3.51	33,612	84
32	DeepSeek V4.1 Flash 1017 High	45.6%	$0.77	385,196	186
33	Muse Wonder Medium	45.2%	$2.82	27,124	75
34	Fable 5.1 Low	45.1%	$5.44	34,795	51
35	DeepSeek V4.1 Flash 1017 Low	44.9%	$0.60	290,651	161
36	Opus 5 High	44.7%	$9.00	61,405	86
37	Haiku 5.5 Extra High	44.3%	$0.56	143,813	93
38	Grok 4.7 High	43.9%	$4.69	56,382	71
39	Opus 5.5 Low	43.7%	$1.17	15,811	28
40	Opus 5 Medium	43.3%	$6.94	45,272	72
41	GLM 5.3 Max	42.6%	$5.05	96,387	166
42	Haiku 5.5 High	42.3%	$0.32	77,057	59
43	GPT-5.6 Sol Max	41.7%	$8.23	42,944	99
44	Muse Wonder Low	41.7%	$1.91	18,880	56
45	Grok 4.7 Medium	41.6%	$3.49	36,683	60
46	Muse Spark 1.3 Max	41.6%	$2.64	52,005	98
47	Grok 4.6 Extra High	41.4%	$6.10	49,814	56
48	GPT-5.6 Terra Max	41.3%	$5.14	60,814	107
49	Opus 5 Low	40.7%	$4.87	31,995	57
50	Grok 4.6 High	40.4%	$5.20	41,387	48
51	Gemini 3.8 Flash High	39.6%	$4.70	162,565	324
52	Sonnet 5.5 Medium	39.2%	$0.52	16,036	22
53	GLM 5.3 High	38.0%	$3.24	60,031	114
54	GPT-5.6 Sol Extra High	37.7%	$4.40	24,729	55
55	Muse Spark 1.3 Extra High	37.5%	$2.10	40,891	83
56	Gemini 3.8 Flash Medium	37.3%	$4.06	128,364	290
57	Haiku 5.5 Medium	36.9%	$0.17	42,659	39
58	GLM 5.3 Flash Max	36.8%	$0.39	56,410	118
59	Grok 4.6 Medium	36.1%	$3.48	24,893	40
60	GPT-5.6 Luna Max	35.9%	$1.03	87,284	208
61	Sonnet 5.5 Low	35.8%	$0.37	11,668	18
62	GPT-5.6 Sol High	35.7%	$2.85	16,174	41
63	Muse Wonder Minimal	34.2%	$0.84	9,761	33
64	Sonnet 5 Max	34.1%	$7.17	149,257	140
65	GPT-5.6 Terra Extra High	33.6%	$1.81	23,436	43
66	Grok 4.6 Low	33.4%	$2.25	16,307	32
67	Muse Spark 1.3 High	33.4%	$1.66	30,654	69
68	GLM 5.3 Low	33.3%	$2.04	31,983	81
69	Grok 4.7 Low	33.1%	$1.58	15,677	40
70	GPT-5.6 Luna Extra High	33.0%	$0.44	40,598	98
71	Muse Spark 1.3 Medium	32.6%	$1.49	27,255	64
72	Sonnet 5 Extra High	32.0%	$4.55	83,373	102
73	GPT-5.6 Sol Medium	31.1%	$1.77	10,111	32
74	GLM 5.3 Flash High	31.1%	$0.25	35,104	84
75	Haiku 5.5 Low	30.9%	$0.08	23,382	25
76	Sonnet 5 High	30.8%	$3.48	61,146	85
77	GPT-5.6 Terra High	30.7%	$1.11	13,162	33
78	GPT-5.6 Luna High	29.4%	$0.25	23,368	64
79	Muse Spark 1.3 Low	29.3%	$0.93	17,483	47
80	Sonnet 5 Medium	28.0%	$2.31	39,114	65
81	Composer 2.5	27.7%	$0.68	17,347	41
82	GPT-5.6 Terra Medium	27.6%	$0.64	7,307	25
83	GLM 5.3 Flash Low	26.9%	$0.15	17,831	58
84	GPT-5.6 Terra Low	25.2%	$0.52	5,914	23
85	GPT-5.6 Sol Low	24.6%	$0.87	4,885	21
86	Muse Spark 1.3 Minimal	24.3%	$0.56	10,620	34
87	Sonnet 5 Low	24.1%	$1.39	23,772	46
88	GPT-5.6 Luna Medium	22.2%	$0.08	7,642	32
89	GPT-5.6 Luna Low	16.0%	$0.03	3,288	18

Changelog

Oct 7, 2026

Reporting

Updated Sonnet 5.5 results to account for adjusted pricing.

Sep 10, 2026

Tasks

CursorBench 4.0
- Introduced new long-horizon problems focused on edit, refactor, investigation, intent understanding, managing jobs, and design adherence.

Aug 11, 2026

Reporting

Updated Sonnet 5 results to account for adjusted pricing.

Jul 30, 2026

Reporting

Updated GPT-5.6 Terra and Luna results to account for adjusted pricing.

Jul 9, 2026

Reporting

Updated GPT-5.6 Sol, Terra, and Luna results to account for cache write costs.

Jul 8, 2026

Tasks

CursorBench 3.2
- Introduced instruction following and advanced tool use problems.

May 19, 2026

Tasks

CursorBench 3.1
- Introduced problems focused on codebase understanding, bugfinding, planning, and code review.
- Improved grading criteria for some edit tasks.

Mar 11, 2026

Tasks

CursorBench 3.0
- Initial set of tasks focused on edit, refactor, and bugfix problems.

CursorBench 4.0

We evaluate agents on ambiguous, multi-file tasks from real Cursor sessions. Higher scores are better.

More about CursorBench

	Model
1	Gemini 4 Carbon High	60.4%	$9.22	39,960	71
2	Gemini 4 Carbon Medium	60.1%	$8.49	37,067	67
3	Fable 5.2 Max	60.0%	$32.79	352,794	218
4	Gemini 4 Carbon Low	58.6%	$11.89	42,663	124
5	Opus 5.5 Max	57.8%	$13.43	218,363	185
6	Fable 5.2 High	56.9%	$9.70	99,259	69
7	Fable 5.2 Extra High	56.3%	$15.48	161,251	108
8	Opus 5.5 Extra High	56.0%	$6.98	101,083	109
9	Opus 5.5 High	56.0%	$3.97	53,078	68
10	Sonnet 5.5 Max	55.5%	$7.05	271,920	170
11	Fable 5.2 Medium	54.8%	$4.66	44,103	35
12	Fable 5.2 Low	53.2%	$3.94	36,616	30
13	Sonnet 5.5 Extra High	53.1%	$2.81	100,158	78
14	Gemini 4 Argon High	52.9%	$4.57	109,500	118
15	Gemini 4 Argon Medium	52.8%	$4.39	103,600	115
16	Opus 5.5 Medium	52.5%	$2.91	37,954	54
17	Fable 5.1 Max	51.8%	$17.28	117,236	128
18	Fable 5.1 Extra High	51.6%	$13.01	87,294	101
19	Gemini 4 Argon Low	50.6%	$4.09	94,200	110
20	Muse Wonder Max	50.4%	$5.25	49,692	114
21	Fable 5.1 High	49.2%	$9.08	58,438	77
22	Haiku 5.5 Max	48.4%	$1.12	325,934	163
23	Muse Wonder Extra High	48.2%	$4.24	39,564	96
24	Sonnet 5.5 High	47.8%	$1.20	37,391	41
25	DeepSeek V4.1 Flash 1017 Max	47.7%	$0.98	537,047	187
26	Fable 5.1 Medium	46.8%	$7.05	45,411	63
27	Opus 5 Max	46.6%	$11.95	85,384	106
28	Grok 4.7 Extra High	46.3%	$6.01	70,141	88
29	DeepSeek V4.1 Flash 1017 Medium	46.3%	$0.80	398,255	189
30	Opus 5 Extra High	46.1%	$11.43	80,094	103
31	Muse Wonder High	46.1%	$3.51	33,612	84
32	DeepSeek V4.1 Flash 1017 High	45.6%	$0.77	385,196	186
33	Muse Wonder Medium	45.2%	$2.82	27,124	75
34	Fable 5.1 Low	45.1%	$5.44	34,795	51
35	DeepSeek V4.1 Flash 1017 Low	44.9%	$0.60	290,651	161
36	Opus 5 High	44.7%	$9.00	61,405	86
37	Haiku 5.5 Extra High	44.3%	$0.56	143,813	93
38	Grok 4.7 High	43.9%	$4.69	56,382	71
39	Opus 5.5 Low	43.7%	$1.17	15,811	28
40	Opus 5 Medium	43.3%	$6.94	45,272	72
41	GLM 5.3 Max	42.6%	$5.05	96,387	166
42	Haiku 5.5 High	42.3%	$0.32	77,057	59
43	GPT-5.6 Sol Max	41.7%	$8.23	42,944	99
44	Muse Wonder Low	41.7%	$1.91	18,880	56
45	Grok 4.7 Medium	41.6%	$3.49	36,683	60
46	Muse Spark 1.3 Max	41.6%	$2.64	52,005	98
47	Grok 4.6 Extra High	41.4%	$6.10	49,814	56
48	GPT-5.6 Terra Max	41.3%	$5.14	60,814	107
49	Opus 5 Low	40.7%	$4.87	31,995	57
50	Grok 4.6 High	40.4%	$5.20	41,387	48
51	Gemini 3.8 Flash High	39.6%	$4.70	162,565	324
52	Sonnet 5.5 Medium	39.2%	$0.52	16,036	22
53	GLM 5.3 High	38.0%	$3.24	60,031	114
54	GPT-5.6 Sol Extra High	37.7%	$4.40	24,729	55
55	Muse Spark 1.3 Extra High	37.5%	$2.10	40,891	83
56	Gemini 3.8 Flash Medium	37.3%	$4.06	128,364	290
57	Haiku 5.5 Medium	36.9%	$0.17	42,659	39
58	GLM 5.3 Flash Max	36.8%	$0.39	56,410	118
59	Grok 4.6 Medium	36.1%	$3.48	24,893	40
60	GPT-5.6 Luna Max	35.9%	$1.03	87,284	208
61	Sonnet 5.5 Low	35.8%	$0.37	11,668	18
62	GPT-5.6 Sol High	35.7%	$2.85	16,174	41
63	Muse Wonder Minimal	34.2%	$0.84	9,761	33
64	Sonnet 5 Max	34.1%	$7.17	149,257	140
65	GPT-5.6 Terra Extra High	33.6%	$1.81	23,436	43
66	Grok 4.6 Low	33.4%	$2.25	16,307	32
67	Muse Spark 1.3 High	33.4%	$1.66	30,654	69
68	GLM 5.3 Low	33.3%	$2.04	31,983	81
69	Grok 4.7 Low	33.1%	$1.58	15,677	40
70	GPT-5.6 Luna Extra High	33.0%	$0.44	40,598	98
71	Muse Spark 1.3 Medium	32.6%	$1.49	27,255	64
72	Sonnet 5 Extra High	32.0%	$4.55	83,373	102
73	GPT-5.6 Sol Medium	31.1%	$1.77	10,111	32
74	GLM 5.3 Flash High	31.1%	$0.25	35,104	84
75	Haiku 5.5 Low	30.9%	$0.08	23,382	25
76	Sonnet 5 High	30.8%	$3.48	61,146	85
77	GPT-5.6 Terra High	30.7%	$1.11	13,162	33
78	GPT-5.6 Luna High	29.4%	$0.25	23,368	64
79	Muse Spark 1.3 Low	29.3%	$0.93	17,483	47
80	Sonnet 5 Medium	28.0%	$2.31	39,114	65
81	Composer 2.5	27.7%	$0.68	17,347	41
82	GPT-5.6 Terra Medium	27.6%	$0.64	7,307	25
83	GLM 5.3 Flash Low	26.9%	$0.15	17,831	58
84	GPT-5.6 Terra Low	25.2%	$0.52	5,914	23
85	GPT-5.6 Sol Low	24.6%	$0.87	4,885	21
86	Muse Spark 1.3 Minimal	24.3%	$0.56	10,620	34
87	Sonnet 5 Low	24.1%	$1.39	23,772	46
88	GPT-5.6 Luna Medium	22.2%	$0.08	7,642	32
89	GPT-5.6 Luna Low	16.0%	$0.03	3,288	18

Changelog

Oct 7, 2026

Reporting

Updated Sonnet 5.5 results to account for adjusted pricing.

Sep 10, 2026

Tasks

CursorBench 4.0
- Introduced new long-horizon problems focused on edit, refactor, investigation, intent understanding, managing jobs, and design adherence.

Aug 11, 2026

Reporting

Updated Sonnet 5 results to account for adjusted pricing.

Jul 30, 2026

Reporting

Updated GPT-5.6 Terra and Luna results to account for adjusted pricing.

Jul 9, 2026

Reporting

Updated GPT-5.6 Sol, Terra, and Luna results to account for cache write costs.

Jul 8, 2026

Tasks

CursorBench 3.2
- Introduced instruction following and advanced tool use problems.

May 19, 2026

Tasks

CursorBench 3.1
- Introduced problems focused on codebase understanding, bugfinding, planning, and code review.
- Improved grading criteria for some edit tasks.

Mar 11, 2026

Tasks

CursorBench 3.0
- Initial set of tasks focused on edit, refactor, and bugfix problems.