Artificial Analysis Omniscience Accuracy

An Artificial Analysis knowledge metric for the proportion of correctly answered questions.

In index
Knowledge

Models176

Top result67.2%Claude Fable 5.1

Top-3 spread1.8 pts

YearN/A

Result and price

Results

176 results

#ModelResult

01

Claude Fable 5.167.2%

02

Claude Opus 5.566.2%

03

Claude Fable 565.4%

04

GPT-6 Astra62.6%

05

Claude Opus 560.9%

06

GPT-5.6 Sol59.4%

07

GPT-5.558.0%

08

Gemini 3 Pro55.8%

09

Gemini 3.7 Flash55.3%

10

Gemini 3.1 Pro54.9%

11

Gemini 3.8 Flash54.6%

12

GPT-6 Sol54.5%

13

GPT-5.3 Codex52.9%

14

Muse Spark 1.152.1%

15

Gemini 3.5 Flash51.9%

16

Grok 4.551.6%

17

GPT-5.450.8%

18

Gemini 3.6 Flash50.0%

19

Muse Spark49.6%

20

DeepSeek V4 Pro 081349.1%

21

Claude Opus 4.7 (Adaptive)48.9%

22

Claude Opus 4.848.8%

23

Grok 4.648.2%

24MAKimi K347.6%

25

Grok 4.747.4%

26

Claude Opus 4.6 (Adaptive)47.0%

27

GPT-5.6 Terra46.8%

28

Claude Opus 4.5 Thinking46.6%

29

DeepSeek V4.1 Flash46.4%

30

Claude Opus 4.645.8%

30

Gemini 3 Flash45.8%

32

Muse Spark 1.245.4%

33

Claude Opus 4.744.7%

34

GPT-5.244.3%

35

GPT-6 Luna43.8%

36

Muse Spark 1.343.6%

37

GPT-5.6 Luna42.7%

38TMInkling41.6%

39STStep 5 Preview41.5%

40

GPT-5.2-Codex41.1%

41

Claude Opus 4.540.9%

42

Grok 440.5%

43

DeepSeek V4 Flash 073140.4%

44

GPT-5 (high)40.3%

45

Claude Sonnet 540.1%

46

GPT-5.1-Codex-Max39.9%

46

GPT-5.1-Codex39.9%

48MAKimi K2.7 Code39.6%

49

GPT-5 (medium)39.5%

50

Gemini 2.5 Pro39.1%

51

Claude Sonnet 4.638.6%

51

o338.6%

53

Qwen 3.6 Max (preview)37.9%

54

GPT-5.137.7%

55

GPT-5.4 mini37.5%

56MAKimi K2.5 (Reasoning)35.2%

56MAKimi K2.535.2%

58

MiMo-V2.6-Pro34.8%

59

Grok 4.334.6%

60

o134.5%

61

GLM-5.333.9%

62TMInkling-Small33.2%

63MAKimi K2.632.6%

64TEHy332.0%

65

Qwen3.8 Max Preview31.7%

65APApodex 1.131.7%

65APApodex 1.1 Mini31.7%

68TEHy3 Preview31.5%

69

Qwen3.7 Max31.1%

70

DeepSeek-R130.5%

71

Gemini 3.5 Flash-Lite29.5%

72

GLM-4.729.3%

72

GLM-5V-Turbo29.3%

74

DeepSeek V3.1 (Reasoning)29.0%

75

GLM-5-Turbo28.4%

76

GPT-4.127.8%

77MAKimi K227.4%

78

Muse Glimmer 30B27.0%

79

MiniMax M2.726.8%

80

MiMo-V2-Pro26.6%

81

Qwen3.6 Plus26.4%

82

GLM-526.3%

83

Gemini 2.5 Flash26.1%

84STStep 3.7 Flash25.8%

85

GPT-5.4 nano25.7%

86

DeepSeek V325.5%

87

Grok 4.1 Fast (Reasoning)25.1%

88

Mistral Large 325.0%

89

Llama 4 Maverick24.9%

90

Mistral Medium 3.5 128B24.7%

91

Qwen3.8-Flash-Next24.5%

91

Qwen3.5 397B24.5%

91

Qwen3.5 397B (Reasoning)24.5%

94

Qwen3.5-122B-A10B24.4%

94

Qwen3 Max24.4%

96

GLM-5.224.3%

96

Nemotron 3 Super 100B24.3%

98

DeepSeek V3.224.0%

99

GLM-5.123.7%

100

Grok Code Fast 123.5%

101

Llama 3.1 405B23.2%

102

DeepSeek V3.123.1%

103

Grok 4 Fast (Reasoning)22.8%

104

Claude 4 Sonnet22.7%

105

Qwen3.7 Plus22.5%

105AATrinity-Large-Thinking22.5%

105AATrinity-Large-Preview22.5%

108

MiMo-V2.5-Pro22.4%

109INMercury 2.522.0%

110

GPT-OSS 120B21.8%

111

Mistral Small 421.7%

—

Mistral Small 4 (Reasoning)21.7%

112

Nemotron 3 Ultra21.6%

113

GLM-4.621.4%

114

Qwen3.5-27B20.7%

115

GPT-4.1 mini20.3%

116

Qwen3.5-35B-A3B20.1%

116

Nemotron Ultra 253B20.1%

118

Gemma 4 31B20.0%

119

GPT-4o19.9%

119

Mistral Large 219.9%

121

Qwen3.6-27B19.6%

122

MiMo-V2-Omni19.3%

123

Gemma 4 26B A4B19.1%

124FAUltravox v0.6 Llama 3.3 70B19.0%

125UPSolar Pro 418.9%

125CONorth Mini Code18.9%

127

Qwen3.6-35B-A3B18.8%

128STA.X K218.6%

129UPSolar Pro 318.5%

130

Mistral Medium 318.3%

131INLing 3.0 Flash18.2%

131INLing 3.0 Flash FP818.2%

133

Claude 3 Haiku17.6%

133SASarvam 105B17.6%

135

Nemotron 3 Nano 30B17.3%

136

Grok 4.1 Fast17.2%

137AMNova Pro16.9%

138

MiniMax M316.7%

139LAK-Exaone16.4%

140

GLM-4.5-Air16.3%

141UPSolar Pro 216.1%

142

GPT-OSS 20B16.0%

143

Qwen3.8-27B15.6%

143

MiMo-V2-Flash15.6%

143

Gemma 4 12B15.6%

143INLing 2.6 Flash15.6%

147MCQuasar 438B15.5%

148

Nemotron 3 Nano Omni 30B A3B15.2%

148

Llama 4 Scout15.2%

150

Qwen3-Omni-30B-A3B-Thinking14.6%

151INLing 3.0 Flash VL14.4%

151

Nemotron 3.5 Lightning 30B A3B NVFP414.4%

153

Qwen3-Omni-30B-A3B-Instruct14.3%

154

Phi-414.1%

155

GPT-4.1 nano13.7%

156LAK-EXAONE 2.013.1%

157

Gemma 3 27B13.0%

158SASarvam 30B12.6%

159IBGranite 4.2 8B11.2%

160CECeleris-111.0%

161LAExaone 4.0 32B10.6%

162IBGranite 4.2 30B10.1%

163LILFM2.5-8B-A1B9.4%

164IBGranite 4.2 3B9.2%

165COCommand A+8.9%

166

Gemma 4 E4B8.6%

167INLing 3.0 Tiny8.5%

168OPMiniCPM5-2B8.4%

169

Gemma 4 E2B6.6%

170LILFM2.5-VL-1.6B-Extract5.8%

171IBGranite-4.0-H-1B5.2%

172LAExaone 4.0 1.2B5.0%

173LILFM2.5-2.6B4.4%

174IBGranite-4.0-350M3.9%

175IBGranite-4.0-H-350M3.8%