SWE-bench Lite Leaderboard

Software Engineering benchmark testing ability to resolve real GitHub issues.

Data from LayerLens

As of November 21, 2026, the top-scoring model on SWE-bench Lite is Claude Opus 4.6 at 62.7%, followed by Claude Opus 4.6 at 62.7% and MiniMax M2.5 at 56.3%. 47 models have been evaluated on this benchmark.

Last updated: November 21, 2026

8 Ways to Use Fewer Tokens

Models

47

Best Score

62.7

Average

27.5

Std Dev

19.5

Categories
Multi-turn
Sponsored The Grid | Spot Priced LLM API
The Grid | Spot Priced LLM API — Why pay list price when providers will bid for your API usage? Up to 80% off.
View spot prices
Provider
Model
Input $/M
Output $/M
SWE-bench Lite
Actions
$5.000
$25.000
62.7
$5.000
$25.000
62.7
$0.270
$0.950
56.3
$1.250
$10.000
54.3
$1.000
$5.000
54.3
$1.000
$5.000
54.3
$0.600
$1.920
53.3
$0.600
$1.920
53.3
$5.000
$25.000
49.3
$5.000
$25.000
49.3
$0.300
$0.900
44.7
$0.550
$2.200
42.0
$0.430
$1.750
42.0
$0.430
$1.750
42.0
$1.000
$10.000
40.0
$0.125
$1.000
38.3
$0.440
$2.200
36.5
$0.000
$0.000
36.3
$0.625
$5.000
36.3
$0.500
$1.500
33.3
$0.100
$0.625
32.3
$0.230
$0.960
29.3
$0.014
$0.028
29.1
$0.300
$2.500
26.1
$0.300
$2.500
26.1
$0.300
$0.900
26.0
$0.300
$0.900
26.0
$0.390
$0.900
20.0
$0.064
$0.224
16.3
$0.064
$0.224
16.3
$0.250
$0.950
14.3
$0.250
$0.950
14.3
$0.500
$3.000
12.7
$0.500
$3.000
12.7
$0.075
$0.400
10.7
$0.030
$0.100
9.0
$0.200
$0.600
8.0
$0.600
$2.200
7.7
$0.075
$0.200
5.7
$0.100
$0.300
4.0
$0.800
$3.200
2.7
$0.000
$0.000
-
$0.270
$0.410
-
$0.270
$0.410
-
$0.490
$2.500
-
$0.490
$2.500
-
$0.260
$1.560
-

Pricing from OpenRouter. Benchmarks from Artificial Analysis.

About SWE-bench Lite

Software Engineering benchmark testing ability to resolve real GitHub issues.

This leaderboard shows all models with SWE-bench Lite benchmark scores, ranked from highest to lowest. Pricing data is included to help you compare performance against cost.

Frequently Asked Questions

Software Engineering benchmark testing ability to resolve real GitHub issues.
As of November 21, 2026, Claude Opus 4.6 leads the SWE-bench Lite leaderboard with a score of 62.7. Rankings change as new models are released and evaluated.
Currently 47 models have been evaluated on SWE-bench Lite, with an average score of 27.5 and standard deviation of 19.5.
Benchmark scores are updated when new evaluations are published by our data sources (Artificial Analysis and LayerLens). Pricing data is refreshed daily from OpenRouter.