Skip to content
LLM Pulse

Leaderboard

SWE-Bench Pro

56 models scored · metric: resolve rate. Prices are the cheapest listed across all serving providers. Pts/$ = score ÷ blended price (3×input + output) / 4 — how much score each dollar buys.

Best value · top 5 by pts per dollar
  1. 1.Laguna XS 2.1635
  2. 2.GPT-5.6 Luna279
  3. 3.Step 3.7 Flash135
  4. 4.GPT-5.4 nano128
  5. 5.LongCat-2.0113
#ModelScoreBest in /MBest out /MPts / $Context
🥇Claude Fable 5anthropic
80.3
$10$504.01M
🥈Claude Opus 5anthropic
79.2
$5$257.91M
🥉Claude Opus 4.8anthropic
69.2
$5$256.91M
4Grok 4.5xai
64.7
$2$621.6500K
5GPT-5.6 Solopenai
64.6
$2.50$1511.51.05M
6Claude Opus 4.7anthropic
64.3
$5$256.41M
7GPT-5.6 Terraopenai
63.4
$1$628.21.05M
8Claude Sonnet 5anthropic
63.2
$2$1015.81M
9GPT-5.6 Lunaopenai
62.7
$0.1$0.62791.05M
10GLM-5.2zhipuai
62.1
$1.40$4.4028.91M
11Muse Spark 1.1meta
61.5
$1.25$4.2530.81M
12Qwen3.7 Maxalibaba
60.6
$2.50$7.5016.21M
13LongCat-2.0meituan
59.5
$0.3$1.201131M
14GPT-5.4openai
59.1
$2.20$1411.51.05M
15MiniMax-M3minimax
59
$0.3$1.20112512K
16Gemini 3.6 Flashgoogle
58.7
$0.75$3.7539.11.05M
17GPT-5.5openai
58.6
$4.55$27.275.71.05M
18MiMo-V2.5-Proxiaomi
57.2
$0.435$0.871051.05M
19Step 3.7 Flashstepfun
56.3
$0.185$1.11135256K
20MiniMax-M2.7minimax
56.2
$0.3$1.20107205K
21Gemini 3.5 Flashgoogle
55.1
$1.50$916.31.05M
22GPT-5.4 miniopenai
54.4
$0.68$436.0400K
23Gemini 3.1 Pro Previewgoogle
54.2
$2$1212.01.05M
24Gemini 3.5 Flash Litegoogle
54.2
$0.3$2.5063.81.05M
25GPT-5.4 nanoopenai
52.4
$0.18$1.10128400K
26Claude Opus 4.6anthropic
51.9
$5$255.21M
27Muse Glimmer 30Bmeta
51.2
$0.3$1.10102131K
28Laguna XS 2.1poolside
47.6
$0.06$0.12635262K
29Gemini 3.1 Pro Previewgoogle
46.1
$2$1210.21.05M
30Claude Opus 4.5anthropic
45.89
$5$254.6200K
31Claude Opus 4.7anthropic
44.9
$5$254.51M
32Claude Sonnet 4.5 (latest)anthropic
43.6
$3$157.3200K
33Gemini 3 Pro Previewgoogle
43.3
$2$129.61.05M
34GPT-5openai
41.78
$1.10$913.6400K
35GPT-5.2 Codexopenai
41.04
$1.60$139.2400K
36North Mini Codecohere
40.2
256K
37Claude Haiku 4.5 (latest)anthropic
39.45
$1$519.7200K
38Qwen3-Coder 480B-A35B Instructalibaba
38.7
$1.50$7.5012.9262K
39MiniMax-M2.1minimax
36.81
$0.3$1.2070.1205K
40Claude Opus 4.7anthropic
36.4
$5$253.61M
41Gemini 3 Flash Previewgoogle
34.63
$0.5$330.81.05M
42Claude Opus 4.7anthropic
34.4
$5$253.41M
43GPT-5.5openai
30.9
$4.55$27.273.01.05M
44GPT-5.2openai
29.94
$1.60$136.7400K
45Kimi K2.6moonshotai
27.3
$0.5$2.6026.6262K
46GPT-5.5openai
26.2
$4.55$27.272.61.05M
47GPT-5.5openai
24.9
$4.55$27.272.41.05M
48Qwen3 235B-A22Balibaba
21.41
$0.7$2.8017.5131K
49GLM-5.1zhipuai
19.8
$1.40$4.409.2200K
50GPT-5.4openai
18.9
$2.20$143.71.05M
51GPT-5.4openai
18.4
$2.20$143.61.05M
52DeepSeek V4 Prodeepseek
18
$0.435$0.8733.11M
53Gemini 3.1 Pro Previewgoogle
15.1
$2$123.41.05M
54Claude Sonnet 4.6anthropic
14.9
$3$152.51M
55Claude Opus 4.6anthropic
11.8
$5$251.21M
56GLM-4.6zhipuai
9.67
$0.6$2.209.7205K

Cheapest scorer: Laguna XS 2.1 at $0.06 input /M.