DeepResearch Bench II

Opus 5
Opus 5High
56.1%
Fable 5.1
Fable 5.1Max
55.0%
Sonnet 4.6
Sonnet 4.6High
53.3%
Opus 5.5
Opus 5.5Max
53.3%
DeepSeek-V4.1-Flash
DeepSeek-V4.1-FlashMax
53.2%
GLM-5.3
GLM-5.3Max
52.0%
Sonnet 5.5
Sonnet 5.5Max
51.7%
GPT-5.6 Sol
GPT-5.6 SolMedium
51.5%
Qwen3.8-Max
Qwen3.8-MaxxHigh
51.0%
GPT-6 Astra
GPT-6 AstraxHigh
50.1%
GPT-5.6 Luna
GPT-5.6 LunaMax
49.7%
Muse Spark Open
Muse Spark OpenxHigh
49.7%
Grok 4.7
Grok 4.7xHigh
49.0%
GPT-5.6 Terra
GPT-5.6 TerraMedium
47.2%
Grok 4.5
Grok 4.5High
47.1%
Opus 4.7
Opus 4.7High
45.4%
GPT-5.5
GPT-5.5Medium
45.4%
Opus 4.8
Opus 4.8High
44.6%
DeepSeek-V4-Pro
DeepSeek-V4-ProHigh
44.2%
Gemini 3.7 Flash
Gemini 3.7 FlashMedium
43.4%
GPT-6 Luna
GPT-6 LunaMax
42.9%
GPT-6 Sol
GPT-6 SolMax
42.9%
Sonnet 5
Sonnet 5High
39.7%
GPT-5.4
GPT-5.4None
38.4%
Gemini 3.6 Flash
Gemini 3.6 FlashMedium
36.3%
Haiku 5.5
Haiku 5.5Max
34.3%
Muse Wonder 1.0
Muse Wonder 1.0Max
10.8%
0%
10%
20%
30%
40%
50%
60%
70%
80%

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.