GDPval Extended

Opus 5.5
Opus 5.5Max
75.8%
Sonnet 5.5
Sonnet 5.5Max
70.9%
Fable 5.1
Fable 5.1Max
68.7%
Haiku 5.5
Haiku 5.5Max
66.0%
Fable 5
Fable 5Max
65.3%
GPT-6 Astra
GPT-6 AstraMax
64.1%
Opus 5
Opus 5Max
62.9%
Grok 4.6
Grok 4.6xHigh
61.8%
Gemini 3.8 Flash
Gemini 3.8 FlashHigh
60.4%
Opus 4.8
Opus 4.8Max
58.9%
GPT-6 Sol
GPT-6 SolMax
58.7%
DeepSeek-V4.1-Flash
DeepSeek-V4.1-FlashMax
58.6%
Gemini 3.7 Flash
Gemini 3.7 FlashHigh
57.8%
Muse Spark 1.2
Muse Spark 1.2xHigh
57.5%
Kimi K3
Kimi K3Max
56.4%
GPT-5.5
GPT-5.5xHigh
54.5%
Sonnet 5
Sonnet 5Max
54.2%
Opus 4.7
Opus 4.7Max
53.8%
Muse Spark 1.1
Muse Spark 1.1xHigh
52.4%
Gemini 3.6 Flash
Gemini 3.6 FlashHigh
52.2%
GPT-5.6 Sol
GPT-5.6 SolMax • Pro
52.0%
Qwen3.8-Max
Qwen3.8-MaxxHigh
51.1%
Sonnet 4.6
Sonnet 4.6High
51.0%
Grok 4.5
Grok 4.5High
51.0%
GPT-6 Luna
GPT-6 LunaMax
48.9%
GPT-5.4
GPT-5.4xHigh
48.0%
GLM-5.2
GLM-5.2Max
47.3%
GPT-5.6 Terra
GPT-5.6 TerraMax
47.2%
GPT-5.6 Luna
GPT-5.6 LunaMax
44.9%
DeepSeek-V4-Pro
DeepSeek-V4-ProMax
42.0%
Gemini 3.1 Pro
Gemini 3.1 ProHigh
41.9%
Kimi K2.7 Code
Kimi K2.7 CodeHigh
41.0%
Inkling
InklingHigh
39.5%
MiniMax-M3
MiniMax-M3High
38.8%
Nemotron 3 Ultra
Nemotron 3 UltraHigh
38.2%
Qwen3.5
Qwen3.5
35.4%
MiniMax-M2.7
MiniMax-M2.7High
32.4%
Gemini 3.5 Flash
Gemini 3.5 FlashHigh
31.6%
Kimi K2
Kimi K2High • Thinking
26.0%
20%
30%
40%
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.