Skip to content

Benchmark

Multi-SWE-bench C++ Verified

Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.

Models measured
11
Scores recorded
31
Sources
1
IQ weight
1.92

Score distribution

6.1%
Mean
5.4%
Median
4.9%
Std dev
1.6%
10th percentile
11.6%
90th percentile
0%
Min
0%25.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.92 for chat models, measured over 11 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT 5InfCode, 20251110_InfCode_GPT-5· Multi-SWE-bench Experiments data docs25.6%
2Claude Sonnet 3.7MopenHands, 20250329_MopenHands_Claude-3.7-Sonnet· Multi-SWE-bench Experiments data docs14.7%
3Claude 3.5 SonnetMopenHands, 20250329_MopenHands_Claude-3.5-Sonnet(Oct)· Multi-SWE-bench Experiments data docs12.4%
4Claude Sonnet 3.7MSWE-agent, 20250329_MSWE-agent_Claude-3.7-Sonnet· Multi-SWE-bench Experiments data docs11.6%
5Gemini 2.5 ProMopenHands, 20250426_MopenHands_Gemini-2.5-Pro· Multi-SWE-bench Experiments data docs9.3%
6Gemini 2.5 ProMSWE-agent, 20250426_MSWE-agent_Gemini-2.5-Pro· Multi-SWE-bench Experiments data docs8.5%
7DeepSeek V3MopenHands, 20250329_MopenHands_DeepSeek-V3· Multi-SWE-bench Experiments data docs7.8%
8DeepSeek V3MSWE-agent, 20250329_MSWE-agent_DeepSeek-V3· Multi-SWE-bench Experiments data docs7.8%
9Claude 3.5 SonnetMSWE-agent, 20250329_MSWE-agent_Claude-3.5-Sonnet(Oct)· Multi-SWE-bench Experiments data docs7%
10GPT 4oMagentLess, 20250329_MagentLess_GPT-4o-1120· Multi-SWE-bench Experiments data docs7%
11Llama 4 MaverickMSWE-agent, 20250426_MSWE-agent_Llama-4-Maverick· Multi-SWE-bench Experiments data docs7%
12o3 MiniMopenHands, 20250329_MopenHands_OpenAI-o3-mini-high· Multi-SWE-bench Experiments data docs7%
13DeepSeek R1MSWE-agent, 20250329_MSWE-agent_DeepSeek-R1· Multi-SWE-bench Experiments data docs6.2%
14Llama 4 MaverickMagentLess, 20250426_MagentLess_Llama-4-Maverick· Multi-SWE-bench Experiments data docs5.4%
15o1MagentLess, 20250329_MagentLess_OpenAI-o1· Multi-SWE-bench Experiments data docs5.4%
16o3 MiniMSWE-agent, 20250329_MSWE-agent_OpenAI-o3-mini-high· Multi-SWE-bench Experiments data docs5.4%
17DeepSeek R1MopenHands, 20250329_MopenHands_DeepSeek-R1· Multi-SWE-bench Experiments data docs4.7%
18Claude 3.5 SonnetMagentLess, 20250329_MagentLess_Claude-3.5-Sonnet(Oct)· Multi-SWE-bench Experiments data docs3.9%
19GPT 4oMopenHands, 20250329_MopenHands_GPT-4o-1120· Multi-SWE-bench Experiments data docs3.9%
20o1MopenHands, 20250329_MopenHands_OpenAI-o1· Multi-SWE-bench Experiments data docs3.9%
21o1MSWE-agent, 20250329_MSWE-agent_OpenAI-o1· Multi-SWE-bench Experiments data docs3.9%
22Claude Sonnet 3.7MagentLess, 20250329_MagentLess_Claude-3.7-Sonnet· Multi-SWE-bench Experiments data docs3.1%
23DeepSeek R1MagentLess, 20250329_MagentLess_DeepSeek-R1· Multi-SWE-bench Experiments data docs3.1%
24Llama 4 MaverickMopenHands, 20250426_MopenHands_Llama-4-Maverick· Multi-SWE-bench Experiments data docs3.1%
25Gemini 2.5 ProMagentLess, 20250426_MagentLess_Gemini-2.5-Pro· Multi-SWE-bench Experiments data docs2.3%
26GPT 4oMSWE-agent, 20250329_MSWE-agent_GPT-4o-1120· Multi-SWE-bench Experiments data docs2.3%
27Qwen2.5 72B InstructMopenHands, 20250329_MopenHands_Qwen2.5-72B-Instruct· Multi-SWE-bench Experiments data docs2.3%
28DeepSeek V3MagentLess, 20250329_MagentLess_DeepSeek-V3· Multi-SWE-bench Experiments data docs1.6%
29o3 MiniMagentLess, 20250329_MagentLess_OpenAI-o3-mini-high· Multi-SWE-bench Experiments data docs1.6%
30Qwen2.5 72B InstructMagentLess, 20250329_MagentLess_Qwen2.5-72B-Instruct· Multi-SWE-bench Experiments data docs0.775%
31Qwen2.5 72B InstructMSWE-agent, 20250329_MSWE-agent_Qwen2.5-72B-Instruct· Multi-SWE-bench Experiments data docs0%

Data credits: Multi-SWE-bench experiments (multi-swe-bench/experiments), Apache-2.0; resolved-rate fractions converted to percentages (licence)