SWE-rebench Leaderboard

1 min read Original article ↗
1

62.7%± 0.91%

70.0%$2.25

2,120,66090.0% cached

2

61.6%± 0.64%

72.7%$1.84

1,866,49791.6% cached

3

60.4%± 1.37%

71.8%$1.75

1,898,13192.5% cached

4

59.6%± 1.98%

72.7%$1.74

1,878,24893.6% cached

5
OpenAI

gpt-5.5-2026-04-23-medium

58.9%± 0.78%

70.0%$0.98

708,41883.5% cached

6

56.5%± 1.20%

67.3%$2.02

2,479,38795.3% cached

7
OpenAI

gpt-5.4-2026-03-05-medium

54.9%± 1.02%

70.9%$0.60

834,45283.5% cached

8

53.1%± 1.45%

66.4%$1.32

1,526,13594.2% cached

9

53.0%± 0.53%

64.5%$0.23

1,031,65398.7% cached

10

51.3%± 0.55%

63.6%$1.29

2,644,57795.6% cached

11

51.1%± 1.20%

66.4%$0.75

1,545,44580.1% cached

12

50.7%± 0.93%

65.5%$0.94

2,664,00191.8% cached

13

49.5%± 0.98%

61.8%$0.77

1,848,59375.7% cached

14

47.8%± 1.37%

60.9%$1.53

1,828,64993.6% cached

15

46.5%± 1.27%

64.5%$0.61

2,466,97790.4% cached

16

45.6%± 1.27%

67.3%$1.06

6,885,81893.5% cached

17

38.2%± 0.86%

59.1%$0.39

2,256,18286.4% cached

18N/AN/AN/AN/A19N/AN/AN/AN/A20N/AN/AN/AN/A21N/AN/AN/AN/A22N/AN/AN/AN/A23N/AN/AN/AN/A24N/AN/AN/AN/A25N/AN/AN/AN/A26N/AN/AN/AN/A27N/AN/AN/AN/A28N/AN/AN/AN/A29
Mistral

Devstral-2-123B-Instruct-2512

N/AN/AN/AN/A30
Mistral

Devstral-Small-2-24B-Instruct-2512

N/AN/AN/AN/A31N/AN/AN/AN/A32N/AN/AN/AN/A33N/AN/AN/AN/A34N/AN/AN/AN/A35N/AN/AN/AN/A36N/AN/AN/AN/A37
Gemini

gemini-2.5-flash-preview-05-20 no-thinking

N/AN/AN/AN/A38
Gemini

gemini-2.5-flash-preview-05-20 no-thinking

N/AN/AN/AN/A39N/AN/AN/AN/A40N/AN/AN/AN/A41N/AN/AN/AN/A42N/AN/AN/AN/A43N/AN/AN/AN/A44N/AN/AN/AN/A45N/AN/AN/AN/A46N/AN/AN/AN/A47N/AN/AN/AN/A48N/AN/AN/AN/A49N/AN/AN/AN/A50N/AN/AN/AN/A51N/AN/AN/AN/A52N/AN/AN/AN/A53N/AN/AN/AN/A54N/AN/AN/AN/A55N/AN/AN/AN/A56N/AN/AN/AN/A57
OpenAI

gpt-5-mini-2025-08-07-high

N/AN/AN/AN/A58
OpenAI

gpt-5-mini-2025-08-07-medium

N/AN/AN/AN/A59N/AN/AN/AN/A60N/AN/AN/AN/A61
OpenAI

gpt-5.2-2025-12-11-medium

N/AN/AN/AN/A62N/AN/AN/AN/A63N/AN/AN/AN/A64N/AN/AN/AN/A65N/AN/AN/AN/A66N/AN/AN/AN/A67N/AN/AN/AN/A68N/AN/AN/AN/A69N/AN/AN/AN/A70N/AN/AN/AN/A71N/AN/AN/AN/A72N/AN/AN/AN/A73N/AN/AN/AN/A74N/AN/AN/AN/A75N/AN/AN/AN/A76N/AN/AN/AN/A77N/AN/AN/AN/A78
Meta

Llama-4-Maverick-17B-128E-Instruct

N/AN/AN/AN/A79
Meta

Llama-4-Scout-17B-16E-Instruct

N/AN/AN/AN/A80N/AN/AN/AN/A81N/AN/AN/AN/A82N/AN/AN/AN/A83N/AN/AN/AN/A84N/AN/AN/AN/A85N/AN/AN/AN/A86N/AN/AN/AN/A87
Qwen

Qwen2.5-Coder-32B-Instruct

N/AN/AN/AN/A88N/AN/AN/AN/A89
Qwen

Qwen3-235B-A22B no-thinking

N/AN/AN/AN/A90N/AN/AN/AN/A91
Qwen

Qwen3-235B-A22B-Instruct-2507

N/AN/AN/AN/A92
Qwen

Qwen3-235B-A22B-Thinking-2507

N/AN/AN/AN/A93
Qwen

Qwen3-30B-A3B-Instruct-2507

N/AN/AN/AN/A94
Qwen

Qwen3-30B-A3B-Thinking-2507

N/AN/AN/AN/A95N/AN/AN/AN/A96N/AN/AN/AN/A97N/AN/AN/AN/A98
Qwen

Qwen3-Coder-30B-A3B-Instruct

N/AN/AN/AN/A99
Qwen

Qwen3-Coder-480B-A35B-Instruct

N/AN/AN/AN/A100N/AN/AN/AN/A101
Qwen

Qwen3-Next-80B-A3B-Instruct

N/AN/AN/AN/A102N/AN/AN/AN/A103N/AN/AN/AN/A104N/AN/AN/AN/A105N/AN/AN/AN/A