SWE-bench Multilingual

DeepSeek-V4.1-Flash
DeepSeek-V4.1-FlashMax
98.2%
DeepSeek-V4-Pro-0813
DeepSeek-V4-Pro-0813Max
96.4%
Opus 5
Opus 5Max
96.2%
Opus 5.5
Opus 5.5Max
95.5%
DeepSeek-V4-Flash
DeepSeek-V4-FlashMax
95.4%
Fable 5
Fable 5Max
94.8%
GPT-5.6 Terra
GPT-5.6 TerraMax
94.4%
GPT-5.6 Sol
GPT-5.6 SolMax • Pro
94.2%
Grok 4.6
Grok 4.6xHigh
92.7%
GPT-5.6 Luna
GPT-5.6 LunaMax
92.4%
Kimi K3
Kimi K3Max
91.4%
Fable 5.1
Fable 5.1Max
91.1%
Sonnet 5.5
Sonnet 5.5Max
90.9%
Sonnet 5
Sonnet 5Max
89.2%
GLM-5.3-Flash
GLM-5.3-FlashMax
88.9%
Qwen3.8-Max
Qwen3.8-MaxxHigh
88.3%
Grok 4.5
Grok 4.5High
87.7%
GLM-5.3
GLM-5.3Max
87.6%
Opus 4.8
Opus 4.8Max
86.6%
Gemini 3.8 Flash
Gemini 3.8 FlashHigh
85.8%
Opus 4.7
Opus 4.7Max
85.3%
Haiku 5.5
Haiku 5.5Max
85.1%
GPT-6 Astra
GPT-6 AstraxHigh
84.9%
GPT-6.1 Sol
GPT-6.1 SolMax
84.9%
Grok 4.7
Grok 4.7xHigh
84.2%
Gemini 3.7 Flash
Gemini 3.7 FlashHigh
82.6%
GPT-6 Sol
GPT-6 SolMax
81.7%
GLM-5.2
GLM-5.2Max
78.6%
GPT-6 Luna
GPT-6 LunaMax
78.2%
Sonnet 4.6
Sonnet 4.6High
78.1%
GPT-5.5
GPT-5.5xHigh
77.7%
Inkling
InklingHigh
77.7%
GPT-5.4
GPT-5.4xHigh
77.0%
MiniMax-M3
MiniMax-M3High
76.0%
Kimi K2.7 Code
Kimi K2.7 CodeHigh
75.6%
Gemini 3.6 Flash
Gemini 3.6 FlashHigh
75.1%
Gemini 3.5 Flash
Gemini 3.5 FlashHigh
68.2%
Gemini 3.1 Pro
Gemini 3.1 ProHigh
65.1%
Qwen3.5
Qwen3.5
61.7%
MiniMax-M2.7
MiniMax-M2.7High
60.8%
DeepSeek-V3.2
DeepSeek-V3.2
60.7%
50%
60%
70%
80%
90%
100%

APEX NEWSLETTER

The latest on frontier AI performance, straight to your inbox.

New benchmarks, leaderboard shifts, and research from the APEX team.

By subscribing, you agree to receive updates from Mercor.
Unsubscribe anytime.