Benchmarks / Artificial Analysis

Reported by Artificial Analysis

Artificial Analysis

Share of expert-written exam questions answered correctly, as run by Artificial Analysis with its own harness and prompts.

Last updated 20 Oct 2026

Results dated
20 Oct 2026
Results
411 configurations of 248 models
Unit
% of questions
Licence
Artificial Analysis commercial data licence

Humanity's Last Exam: Beam 1.0 501B A23B

Top 15 of 411 results · % of questions, higher is better. Choose a model to highlight it.Clear highlight

  1. 1 Claude Opus 5.5 (max reasoning)Anthropic 61.4%
  2. 2 Claude Fable 5.1 (max reasoning)Anthropic 59.1%
  3. 3 Claude Fable 5.1 (extra-high reasoning)Anthropic 58.7%
  4. 4 Claude Opus 5.5 (extra-high reasoning)Anthropic 57.5%
  5. 5 Gemini 4 Argon (high reasoning)Google 57.1%
  6. 6 Claude Fable 5.1 (high reasoning)Anthropic 55.9%
  7. 7 Claude Opus 5.5 (high reasoning)Anthropic 55.6%
  8. 8 Claude Fable 5 (max reasoning)Anthropic 55.5%
  9. 9 Claude Sonnet 5.5 (max reasoning)Anthropic 55.0%
  10. 10 Claude Opus 5 (max reasoning)Anthropic 54.9%
  11. 11 Claude Opus 5.5 (medium reasoning)Anthropic 54.7%
  12. 11 GPT-6 Astra (max reasoning)OpenAI 54.7%
  13. 13 GPT-6 Astra (extra-high reasoning)OpenAI 54.6%
  14. 14 Claude Opus 5 (extra-high reasoning)Anthropic 54.4%
  15. 15 Claude Fable 5.1 (medium reasoning)Anthropic 53.8%
  16. 119 Beam 1.0 501B A23BReflection 34.7%

Full results

Artificial Analysis: Humanity's Last Exam, % of questions, higher is better
#ModelHumanity's Last Exam
% of questions, higher is better		Price
$ per million tokens, in / out
1 Claude Opus 5.5 (max reasoning)Anthropic
61.4%
$4 / $20
2 Claude Fable 5.1 (max reasoning)Anthropic
59.1%
$10 / $50
3 Claude Fable 5.1 (extra-high reasoning)Anthropic
58.7%
$10 / $50
4 Claude Opus 5.5 (extra-high reasoning)Anthropic
57.5%
$4 / $20
5 Gemini 4 Argon (high reasoning)Google
57.1%
$2 / $10
6 Claude Fable 5.1 (high reasoning)Anthropic
55.9%
$10 / $50
7 Claude Opus 5.5 (high reasoning)Anthropic
55.6%
$4 / $20
8 Claude Fable 5 (max reasoning)Anthropic
55.5%
$10 / $50
9 Claude Sonnet 5.5 (max reasoning)Anthropic
55.0%
$2 / $10
10 Claude Opus 5 (max reasoning)Anthropic
54.9%
$5 / $25
11 Claude Opus 5.5 (medium reasoning)Anthropic
54.7%
$4 / $20
11 GPT-6 Astra (max reasoning)OpenAI
54.7%
$10 / $50
13 GPT-6 Astra (extra-high reasoning)OpenAI
54.6%
$10 / $50
14 Claude Opus 5 (extra-high reasoning)Anthropic
54.4%
$5 / $25
15 Claude Fable 5.1 (medium reasoning)Anthropic
53.8%
$10 / $50
16 GPT-6 Astra (high reasoning)OpenAI
53.1%
$10 / $50
17 GPT-6.1 Sol (max reasoning)OpenAI
52.9%
$2 / $10
18 Claude Opus 5 (high reasoning)Anthropic
52.8%
$5 / $25
19 GPT-6 Astra (medium reasoning)OpenAI
52.7%
$10 / $50
20 GPT-6.1 Sol (extra-high reasoning)OpenAI
52.6%
$2 / $10
21 GPT-6.1 Sol (high reasoning)OpenAI
51.4%
$2 / $10
22 Claude Opus 5 (medium reasoning)Anthropic
51.3%
$5 / $25
23 Muse Wonder (max reasoning)Meta
50.1%
$4 / $20
24 Claude Sonnet 5.5 (extra-high reasoning)Anthropic
50.0%
$2 / $10
25 GPT-6.1 Sol (medium reasoning)OpenAI
49.9%
$2 / $10
26 GPT-5.6 Sol (max reasoning)OpenAI
49.5%
$4 / $20
27 MiMo-V2.6-ProXiaomi
49.4%
$0.43 / $0.87
28 GPT-6 Astra (low reasoning)OpenAI
49.2%
$10 / $50
29 Claude Fable 5.1 (low reasoning)Anthropic
48.9%
$10 / $50
30 Claude Opus 4.8 (max reasoning)Anthropic
48.7%
$5 / $25
30 Muse Spark 1.3 (max reasoning)Meta
48.7%
$1.25 / $4.25
32 Claude Opus 5.5 (low reasoning)Anthropic
48.3%
$4 / $20
33 MiniMax M3.1 FlashMiniMax
48.0%
$0.15 / $0.60
34 Gemini 3.7 Flash (high reasoning)Google
47.9%
$1.50 / $7.50
34 GPT-6 Sol (max reasoning)OpenAI
47.9%
$2 / $10
36 Gemini 3.8 Flash (high reasoning)Google
47.8%
$1.50 / $7.50
37 Muse Spark 1.3 (extra-high reasoning)Meta
47.5%
$1.25 / $4.25
38 GPT-6.1 Sol (low reasoning)OpenAI
47.4%
$2 / $10
39 GPT-5.6 Sol (extra-high reasoning)OpenAI
47.3%
$4 / $20
40 Gemini 3.1 Pro PreviewGoogle
47.0%
$2 / $12
41 Kimi K3 (max reasoning)Moonshot AI
46.9%
$3 / $15
42 GPT-6 Sol (extra-high reasoning)OpenAI
46.3%
$2 / $10
43 Muse Spark 1.1 (extra-high reasoning)Meta
46.2%
$1.25 / $4.25
44 GPT-5.6 Sol (high reasoning)OpenAI
46.0%
$4 / $20
45 Claude Sonnet 5.5 (high reasoning)Anthropic
45.8%
$2 / $10
45 GPT-5.5 (extra-high reasoning)OpenAI
45.8%
$5 / $30
47 Muse Spark 1.2 (extra-high reasoning)Meta
45.5%
$1.25 / $4.25
48 GPT-5.5 (high reasoning)OpenAI
45.0%
$5 / $30
49 Claude Haiku 5.5 (max reasoning)Anthropic
44.4%
$0.10 / $0.50
50 GPT-6 Sol (high reasoning)OpenAI
44.1%
$2 / $10
50 Grok 4.6 (extra-high reasoning)xAI
44.1%
$2 / $6
52 Muse Spark Open (extra-high reasoning)Meta
43.9%
–
53 GPT-5.4 (extra-high reasoning)OpenAI
43.7%
$2.50 / $15
54 Claude Opus 5 (low reasoning)Anthropic
43.4%
$5 / $25
55 DeepSeek-V4.1-Flash-1017 (max reasoning)DeepSeek
43.3%
$0.30 / $1.20
56 Qwen3.8-Max (0902)Alibaba
43.1%
$2 / $6
56 Grok 4.7 (extra-high reasoning)xAI
43.1%
$2 / $6
58 Qwen3.8-Max (0803)Alibaba
43.0%
–
59 GPT-5.6 Terra (max reasoning)OpenAI
42.9%
$2 / $12
59 Grok 4.6 (high reasoning)xAI
42.9%
$2 / $6
61 Claude Haiku 5.5 (extra-high reasoning)Anthropic
42.7%
$0.10 / $0.50
61 Gemini 3.5 Flash (high reasoning)Google
42.7%
$1.50 / $9
61 Grok 4.5 (high reasoning)xAI
42.7%
$2 / $6
64 GPT-5.3-Codex (extra-high reasoning)OpenAI
42.5%
$1.75 / $14
65 Qwen3.8-2.4T-A95BAlibaba
42.4%
$2 / $6
65 GPT-5.5 (medium reasoning)OpenAI
42.4%
$5 / $30
67 Claude Opus 4.7 (max reasoning)Anthropic
42.3%
$5 / $25
67 Grok 4.7 (high reasoning)xAI
42.3%
$2 / $6
67 GLM-5.3 (max reasoning)Z.ai
42.3%
$1.40 / $4.40
70 GPT-5.6 Sol (medium reasoning)OpenAI
42.2%
$4 / $20
71 Gemini 3.8 Flash (medium reasoning)Google
42.1%
$1.50 / $7.50
71 Grok 4.6 (medium reasoning)xAI
42.1%
$2 / $6
73 GPT-5.6 Terra (extra-high reasoning)OpenAI
41.9%
$2 / $12
74 Claude Sonnet 5 (max reasoning)Anthropic
41.3%
$2 / $10
74 Gemini 3.5 Flash (medium reasoning)Google
41.3%
$1.50 / $9
76 GLM-5.2 (max reasoning)Z.ai
41.1%
$1.40 / $4.40
77 DeepSeek-V4-Pro (0813, max reasoning)DeepSeek
41.0%
$1.32 / $3.96
77 GPT-6 Sol (medium reasoning)OpenAI
41.0%
$2 / $10
79 Gemini 3.6 Flash (high reasoning)Google
40.8%
$1.50 / $7.50
80 Muse SparkMeta
40.7%
–
81 Qwen3.7-MaxAlibaba
40.5%
$1.48 / $4.42
82 Claude Opus 4.6 (max reasoning)Anthropic
39.9%
$5 / $25
82 GLM-5.3-FlashZ.ai
39.9%
$0.15 / $0.50
84 Claude Sonnet 5.5 (medium reasoning)Anthropic
39.8%
$2 / $10
85 Gemini 3 Pro Preview (high reasoning)Google
39.7%
–
86 GPT-5.6 Luna (max reasoning)OpenAI
39.5%
$0.20 / $1.20
87 GPT-5.6 Sol (low reasoning)OpenAI
39.4%
$4 / $20
88 DeepSeek-V4.1-Flash (max reasoning)DeepSeek
39.2%
$0.30 / $1.20
88 Grok 4.7 (low reasoning)xAI
39.2%
$2 / $6
90 Claude Sonnet 5 (extra-high reasoning)Anthropic
39.0%
$2 / $10
90 Gemini 3.7 Flash (medium reasoning)Google
39.0%
$1.50 / $7.50
90 MiniMax-M3MiniMax
39.0%
$0.30 / $1.20
93 DeepSeek-V4-Flash (0731, max reasoning)DeepSeek
38.6%
$0.14 / $0.28
94 GPT-5.6 Terra (high reasoning)OpenAI
38.5%
$2 / $12
94 GPT-6 Luna (max reasoning)OpenAI
38.5%
$0.10 / $0.50
96 Grok Build 0.1xAI
38.3%
$1 / $2
97 Qwen3.8-Flash-NextAlibaba
38.0%
–
98 GPT-5.2 (extra-high reasoning)OpenAI
37.7%
$1.75 / $14
99 DeepSeek-V4-Pro (0423, max reasoning)DeepSeek
37.5%
$1.42 / $2.83
99 Kimi K2.6Moonshot AI
37.5%
$0.95 / $4
101 Claude Haiku 5.5 (high reasoning)Anthropic
37.3%
$0.10 / $0.50
102 Grok 4.3 (high reasoning)xAI
37.2%
$1.25 / $2.50
103 Gemini 3.8 Flash (low reasoning)Google
37.1%
$1.50 / $7.50
104 GPT-5.6 Luna (extra-high reasoning)OpenAI
37.0%
$0.20 / $1.20
105 Gemini 3 Flash Preview (reasoning on)Google
36.6%
$0.50 / $3
105 GLM-5.3 (low reasoning)Z.ai
36.6%
$1.40 / $4.40
107 Claude Sonnet 5.5 (low reasoning)Anthropic
36.2%
$2 / $10
108 Claude Sonnet 5 (high reasoning)Anthropic
35.7%
$2 / $10
108 GPT-5.2-Codex (extra-high reasoning)OpenAI
35.7%
$1.75 / $14
108 MiMo-V2.5-Pro (reasoning on)Xiaomi
35.7%
$0.43 / $0.87
111 Qwen3.7-PlusAlibaba
35.6%
$0.32 / $1.28
112 DeepSeek-V4-Pro (0423, high reasoning)DeepSeek
35.2%
$1.42 / $2.83
113 Gemini 3.7 Flash (low reasoning)Google
35.1%
$1.50 / $7.50
113 MiMo-V2.6-FlashXiaomi
35.1%
$0.14 / $0.28
115 Mistral Large 4Mistral AI
35.0%
$1.36 / $4.18
115 Kimi K2.7 CodeMoonshot AI
35.0%
$0.95 / $4
117 GPT-6 Sol (low reasoning)OpenAI
34.9%
$2 / $10
118 DeepSeek-V4-Flash (0423, max reasoning)DeepSeek
34.8%
$0.14 / $0.28
119 Beam 1.0 501B A23BReflection
34.7%
–
120 DeepSeek-V4-Flash-Vision-Exp (max reasoning)DeepSeek
34.5%
$0.44 / $1.32
120 Grok 4.20xAI
34.5%
$1.25 / $2.50
122 GPT-6 Luna (extra-high reasoning)OpenAI
34.3%
$0.10 / $0.50
123 Qwen3.8-27B (extra-high reasoning)Alibaba
33.9%
$0.50 / $3
124 Claude Haiku 5.5 (medium reasoning)Anthropic
33.8%
$0.10 / $0.50
125 Claude Sonnet 4.6 (max reasoning)Anthropic
33.6%
$3 / $15
126 Hy3Tencent
33.5%
$0.14 / $0.58
127 GPT-5.6 Luna (high reasoning)OpenAI
33.4%
$0.20 / $1.20
128 Claude Opus 4.7 (no reasoning)Anthropic
33.3%
$5 / $25
128 GPT-5.6 Terra (medium reasoning)OpenAI
33.3%
$2 / $12
128 Inkling SmallThinking Machines
33.3%
$0.45 / $1.20
131 GPT-6 Luna (high reasoning)OpenAI
32.9%
$0.10 / $0.50
132 GPT-5.5 (low reasoning)OpenAI
32.7%
$5 / $30
133 Grok 4.20 (0309, reasoning)xAI
32.4%
–
134 Inkling (extra-high reasoning)Thinking Machines
31.9%
$0.95 / $4.05
135 Qwen3.6-Max-PreviewAlibaba
30.8%
$1.03 / $6.16
135 GPT-5.4 (low reasoning)OpenAI
30.8%
$2.50 / $15
137 Kimi K2.5Moonshot AI
30.7%
$0.57 / $2.85
137 Nemotron 3.5 Super VLNVIDIA
30.7%
–
139 DeepSeek-V4-Flash (0423, high reasoning)DeepSeek
30.3%
$0.14 / $0.28
140 Claude Opus 4.5 (reasoning on)Anthropic
30.1%
$5 / $25
140 GLM-5.1Z.ai
30.1%
$1.38 / $4.40
142 Claude Sonnet 5 (medium reasoning)Anthropic
30.0%
$2 / $10
142 Grok 4.3 (medium reasoning)xAI
30.0%
$1.25 / $2.50
144 Solar Pro 4Upstage
29.8%
$0.09 / $0.36
145 MiniMax-M2.7MiniMax
29.6%
$0.30 / $1.20
146 Gemini 3 Pro Preview (low reasoning)Google
29.5%
–
147 GLM-5Z.ai
29.3%
$0.95 / $2.55
148 GPT-5.6 Terra (low reasoning)OpenAI
29.2%
$2 / $12
149 Qwen3.5-397B-A17BAlibaba
29.0%
$0.55 / $3.50
150 DeepSeek-V3.2-SpecialeDeepSeek
28.7%
–
151 GPT-5.1 (high reasoning)OpenAI
28.5%
$1.25 / $10
151 GPT-5 (high reasoning)OpenAI
28.5%
$1.25 / $10
153 GPT-5.4 nano (extra-high reasoning)OpenAI
28.3%
$0.20 / $1.25
153 GPT-6 Luna (medium reasoning)OpenAI
28.3%
$0.10 / $0.50
155 GPT-5.4 mini (extra-high reasoning)OpenAI
28.1%
$0.75 / $4.50
156 Qwen3-Max-ThinkingAlibaba
28.0%
$0.78 / $3.90
157 Grok 4.20 (no reasoning)xAI
27.9%
$1.25 / $2.50
157 GLM-5.1 (no reasoning)Z.ai
27.9%
$1.38 / $4.40
159 Qwen3.6-PlusAlibaba
27.8%
$0.33 / $1.95
159 GPT-5-Codex (high reasoning)OpenAI
27.8%
–
159 Hy3 Preview (reasoning on)Tencent
27.8%
$0.18 / $0.60
159 GLM-5-TurboZ.ai
27.8%
$1.20 / $4
163 Grok 4.6 (low reasoning)xAI
27.6%
$2 / $6
164 GLM-4.7Z.ai
27.4%
$0.54 / $1.98
165 MiMo-V2.5Xiaomi
27.2%
$0.17 / $0.34
166 Claude Haiku 5.5 (low reasoning)Anthropic
27.0%
$0.10 / $0.50
167 GPT-5.2 (medium reasoning)OpenAI
26.7%
$1.75 / $14
167 Grok 4xAI
26.7%
–
169 GPT-5.6 Luna (medium reasoning)OpenAI
25.8%
$0.20 / $1.20
170 GPT-5.1-Codex (high reasoning)OpenAI
25.7%
$1.25 / $10
171 GPT-5 (medium reasoning)OpenAI
25.4%
$1.25 / $10
172 Qwen3.5-122B-A10BAlibaba
25.2%
$0.26 / $2.08
173 Kimi K3 (low reasoning)Moonshot AI
25.0%
$3 / $15
174 DeepSeek-V3.2 (reasoning on)DeepSeek
24.6%
$0.30 / $0.96
175 Grok 4.20 (0309, non-reasoning, no reasoning)xAI
24.5%
–
176 Gemini 3.5 Flash (minimal reasoning)Google
24.1%
$1.50 / $9
177 Qwen3.5-27BAlibaba
23.9%
$0.27 / $2.16
178 Kimi K2 ThinkingMoonshot AI
23.8%
$0.60 / $2.50
179 Gemma 4 31BGoogle
23.6%
$0.14 / $0.40
180 MiniMax-M2.1MiniMax
23.2%
$0.30 / $1.20
181 Qwen3.6-27BAlibaba
23.1%
$0.30 / $3.20
182 Gemini 2.5 ProGoogle
22.5%
$1.25 / $10
183 Qwen3.6-35B-A3BAlibaba
22.2%
$0.10 / $1
184 Muse Glimmer (high reasoning)Meta
22.0%
–
185 Claude Sonnet 5 (low reasoning)Anthropic
21.9%
$2 / $10
186 GPT-5.5 Instant (2026-05-26)OpenAI
21.6%
–
187 GPT-5 mini (high reasoning)OpenAI
21.5%
$0.25 / $2
188 Step 3.5 FlashStepFun
21.1%
$0.10 / $0.30
189 Qwen3.5-35B-A3BAlibaba
21.0%
$0.16 / $1.30
190 Nemotron 3 Super 120B A12BNVIDIA
20.8%
$0.085 / $0.40
191 MiniMax-M2.5MiniMax
20.5%
$0.30 / $1.20
192 GPT-6 Luna (low reasoning)OpenAI
20.3%
$0.10 / $0.50
193 o3OpenAI
20.1%
$2 / $8
194 GPT-5.5 Instant (2026-06-26)OpenAI
19.9%
–
195 Qwen3.5-397B-A17B (no reasoning)Alibaba
19.8%
$0.55 / $3.50
195 GPT-5.6 Luna (low reasoning)OpenAI
19.8%
$0.20 / $1.20
197 Kimi K2.6 (no reasoning)Moonshot AI
19.6%
$0.95 / $4
197 GPT-5 (low reasoning)OpenAI
19.6%
$1.25 / $10
197 gpt-oss-120b (high reasoning)OpenAI
19.6%
$0.15 / $0.60
200 Gemma 4 26B A4BGoogle
19.3%
$0.10 / $0.30
200 Grok 4.1 Fast (reasoning)xAI
19.3%
–
202 Claude Opus 4.6 (no reasoning)Anthropic
19.1%
$5 / $25
202 Grok 4 Fast (reasoning)xAI
19.1%
–
204 Claude Sonnet 5 (no reasoning)Anthropic
19.0%
$2 / $10
205 Gemini 3.5 Flash-LiteGoogle
18.8%
$0.30 / $2.50
206 GPT-5.4 mini (medium reasoning)OpenAI
18.6%
$0.75 / $4.50
207 GPT-5.1-Codex-Mini (high reasoning)OpenAI
18.5%
$0.25 / $2
208 GPT-6 Sol (no reasoning)OpenAI
18.4%
$2 / $10
208 Grok 4.3 (low reasoning)xAI
18.4%
$1.25 / $2.50
210 Claude Sonnet 4.5 (reasoning on)Anthropic
17.8%
$3 / $15
211 Gemini 3.1 Flash-Lite PreviewGoogle
17.2%
$0.25 / $1.50
212 Mercury 2Inception
17.1%
$0.25 / $0.75
212 GLM-5V-TurboZ.ai
17.1%
$1.20 / $4
214 GPT-5.6 Sol (no reasoning)OpenAI
16.7%
$4 / $20
215 o4-mini (high reasoning)OpenAI
16.5%
$1.10 / $4.40
216 DeepSeek-V3.1-Terminus (reasoning on)DeepSeek
16.4%
$0.27 / $1
217 Qwen3-235B-A22B-Thinking-2507Alibaba
15.9%
$0.30 / $3
217 Qwen3.5-122B-A10B (no reasoning)Alibaba
15.9%
$0.26 / $2.08
217 GPT-5.4 nano (medium reasoning)OpenAI
15.9%
$0.20 / $1.25
217 GPT-5 mini (medium reasoning)OpenAI
15.9%
$0.25 / $2
221 Trinity Large ThinkingArcee AI
15.8%
$0.25 / $0.80
221 DeepSeek-R1-0528DeepSeek
15.8%
$0.50 / $2.18
223 Gemma 4 12BGoogle
15.7%
–
224 Qwen3.6-27B (no reasoning)Alibaba
15.1%
$0.30 / $3.20
225 Gemini 3 Flash Preview (no reasoning)Google
15.0%
$0.50 / $3
226 Qwen3.5-9BAlibaba
14.9%
$0.10 / $0.15
226 Qwen3.5-Omni-PlusAlibaba
14.9%
–
226 DeepSeek-V3.2-Exp (reasoning on)DeepSeek
14.9%
$0.27 / $0.41
229 MiMo-V2.5-Pro (no reasoning)Xiaomi
14.8%
$0.43 / $0.87
230 GLM-4.6 (reasoning on)Z.ai
14.5%
$0.50 / $2
231 DeepSeek-V3.1 (reasoning on)DeepSeek
14.3%
$0.55 / $1.65
232 Qwen3.8-27B (medium reasoning)Alibaba
14.1%
$0.50 / $3
233 Qwen3.8-27B (low reasoning)Alibaba
14.0%
$0.50 / $3
234 Qwen3.5-27B (no reasoning)Alibaba
13.9%
$0.27 / $2.16
234 Qwen3.6-35B-A3B (no reasoning)Alibaba
13.9%
$0.10 / $1
236 Gemini 2.5 Flash Preview (09-2025, reasoning on)Google
13.8%
–
236 Mistral Medium 3.5Mistral AI
13.8%
$1.50 / $7.50
238 MiniMax-M2MiniMax
13.7%
$0.30 / $1.20
238 GPT-5.5 (no reasoning)OpenAI
13.7%
$5 / $30
240 Qwen3.5-35B-A3B (no reasoning)Alibaba
13.4%
$0.16 / $1.30
241 Claude Sonnet 4.6 (no reasoning)Anthropic
13.3%
$3 / $15
242 Claude Opus 4.5 (no reasoning)Anthropic
13.2%
$5 / $25
242 Kimi K2.5 (no reasoning)Moonshot AI
13.2%
$0.57 / $2.85
244 GLM-4.5Z.ai
13.0%
$0.60 / $2.20
245 Qwen3-Max-Thinking-PreviewAlibaba
12.7%
–
246 Qwen3-Next-80B-A3B-ThinkingAlibaba
12.6%
$0.15 / $1.20
247 Claude Opus 4.1 (reasoning on)Anthropic
12.5%
$15 / $75
248 Qwen3.8-27B (no reasoning)Alibaba
12.1%
$0.50 / $3
248 Gemini 2.5 Flash (reasoning on)Google
12.1%
$0.30 / $2.50
250 o3-mini (high reasoning)OpenAI
12.0%
$1.10 / $4.40
251 Qwen3-MaxAlibaba
11.9%
$0.78 / $3.90
251 Qwen3-VL-235B-A22B-ThinkingAlibaba
11.9%
$0.40 / $4
253 Gemma 4 31B (no reasoning)Google
11.8%
$0.14 / $0.40
254 Nova 2 Lite (high reasoning)Amazon
11.6%
$0.30 / $2.50
255 Gemma 4 26B A4B (no reasoning)Google
11.5%
$0.10 / $0.30
256 Nemotron 3 Nano 30B A3B (reasoning on)NVIDIA
11.4%
$0.05 / $0.20
256 GPT-5.6 Terra (no reasoning)OpenAI
11.4%
$2 / $12
258 GPT-5.4 (no reasoning)OpenAI
11.3%
$2.50 / $15
259 DeepSeek-V3.2 (no reasoning)DeepSeek
11.2%
$0.30 / $0.96
260 Qwen3-235B-A22B-Instruct-2507Alibaba
11.1%
$0.15 / $0.75
261 Qwen3-235B-A22B (reasoning on)Alibaba
11.0%
$0.46 / $1.82
261 gpt-oss-20b (high reasoning)OpenAI
11.0%
$0.03 / $0.15
263 DeepSeek-V4.1-Flash (no reasoning)DeepSeek
10.8%
$0.30 / $1.20
263 DiffusionGemma 26B A4BGoogle
10.8%
–
265 Claude Sonnet 4 (reasoning on)Anthropic
10.7%
$3 / $15
266 DeepSeek-V4-Pro (0813, no reasoning)DeepSeek
10.6%
$1.32 / $3.96
267 Claude Haiku 4.5 (reasoning on)Anthropic
10.4%
$1 / $5
268 Qwen3-30B-A3B-Thinking-2507Alibaba
10.3%
$0.20 / $2.40
268 Magistral Medium 1.2Mistral AI
10.3%
–
268 Solar Pro 3Upstage
10.3%
$0.15 / $0.60
271 Qwen3-Coder-NextAlibaba
10.1%
$0.18 / $0.90
271 Qwen3-Max-PreviewAlibaba
10.1%
–
271 Qwen3-VL-32B-ThinkingAlibaba
10.1%
–
274 Qwen3.5-4B (reasoning on)Alibaba
9.9%
–
274 Mistral Small 4Mistral AI
9.9%
$0.15 / $0.60
276 GLM-5.2 (no reasoning)Z.ai
9.8%
$1.40 / $4.40
277 Granite 4.2 8BIBM
9.7%
$0.06 / $0.25
278 GLM-4.6V (reasoning on)Z.ai
9.6%
$0.30 / $0.90
279 GPT-5 nano (high reasoning)OpenAI
9.5%
$0.05 / $0.40
280 Qwen3.5-9B (no reasoning)Alibaba
9.4%
$0.10 / $0.15
281 Nova 2 Lite (medium reasoning)Amazon
9.0%
$0.30 / $2.50
281 DeepSeek-V3.2-Exp (no reasoning)DeepSeek
9.0%
$0.27 / $0.41
283 Qwen3-VL-30B-A3B-ThinkingAlibaba
8.9%
$0.29 / $1
284 DeepSeek-V3.1-Terminus (no reasoning)DeepSeek
8.7%
$0.27 / $1
284 Gemini 2.5 Flash Preview (09-2025, no reasoning)Google
8.7%
–
284 GPT-5 nano (medium reasoning)OpenAI
8.7%
$0.05 / $0.40
287 GPT-6 Luna (no reasoning)OpenAI
8.6%
$0.10 / $0.50
288 DeepSeek-R1DeepSeek
8.5%
$0.70 / $2.50
289 DeepSeek-V4-Pro (0423, no reasoning)DeepSeek
8.2%
$1.42 / $2.83
290 Qwen3.5-4B (no reasoning)Alibaba
8.0%
–
290 GPT-5.2 (no reasoning)OpenAI
8.0%
$1.75 / $14
290 Grok Code Fast 1xAI
8.0%
–
293 o3-miniOpenAI
7.9%
$1.10 / $4.40
294 DeepSeek-V4-Flash (0423, no reasoning)DeepSeek
7.8%
$0.14 / $0.28
295 Qwen3.5-Omni-FlashAlibaba
7.6%
–
295 Qwen3-Next-80B-A3B-InstructAlibaba
7.6%
$0.10 / $1.10
295 GLM-4.7-FlashZ.ai
7.6%
$0.06 / $0.40
295 GLM-5 (no reasoning)Z.ai
7.6%
$0.95 / $2.55
299 Qwen3-Omni-30B-A3B-ThinkingAlibaba
7.5%
–
300 Qwen3-32B (reasoning on)Alibaba
7.4%
$0.14 / $0.40
300 Kimi K2 (0711)Moonshot AI
7.4%
$0.57 / $2.30
302 Claude Sonnet 4.5 (no reasoning)Anthropic
7.2%
$3 / $15
302 GPT-5.6 Luna (no reasoning)OpenAI
7.2%
$0.20 / $1.20
304 Gemini 2.5 Flash-Lite Preview (09-2025, reasoning on)Google
7.0%
–
304 o1OpenAI
7.0%
$15 / $60
304 Hy3 Preview (no reasoning)Tencent
7.0%
$0.18 / $0.60
304 GLM-4.5-AirZ.ai
7.0%
$0.14 / $0.86
308 Qwen3-30B-A3B-Instruct-2507Alibaba
6.9%
$0.09 / $0.30
309 Qwen3-VL-32B-InstructAlibaba
6.8%
$0.10 / $0.42
309 Gemini 2.5 Flash-Lite (reasoning on)Google
6.8%
$0.10 / $0.40
309 Grok 4.3 (no reasoning)xAI
6.8%
$1.25 / $2.50
312 DeepSeek-V3.1 (no reasoning)DeepSeek
6.7%
$0.55 / $1.65
313 Qwen3-VL-235B-A22B-InstructAlibaba
6.6%
$0.30 / $1.50
313 GPT-5 ChatOpenAI
6.6%
–
315 Magistral Small 1.2Mistral AI
6.4%
–
315 Kimi K2 (0905)Moonshot AI
6.4%
$0.60 / $2.50
315 GLM-4.7 (no reasoning)Z.ai
6.4%
$0.54 / $1.98
318 Qwen3-VL-30B-A3B-InstructAlibaba
6.3%
$0.15 / $0.60
318 Gemma 4 12B (no reasoning)Google
6.3%
–
318 GLM-4.5V (reasoning on)Z.ai
6.3%
$0.60 / $1.80
321 Qwen3-30B-A3B (reasoning on)Alibaba
6.2%
$0.12 / $0.50
321 Qwen3-4B-Thinking-2507Alibaba
6.2%
–
323 GPT-5 (minimal reasoning)OpenAI
6.0%
$1.25 / $10
324 GPT-5.4 mini (no reasoning)OpenAI
5.9%
$0.75 / $4.50
324 gpt-oss-120b (low reasoning)OpenAI
5.9%
$0.15 / $0.60
326 Llama 3.2 1B InstructMeta
5.5%
$0.027 / $0.20
326 GLM-4.6 (no reasoning)Z.ai
5.5%
$0.50 / $2
328 Ministral 3 3BMistral AI
5.4%
$0.10 / $0.10
329 Gemma 3 4BGoogle
5.3%
$0.05 / $0.10
329 Llama 3.1 8B InstructMeta
5.3%
$0.05 / $0.08
329 Llama 3.2 3B InstructMeta
5.3%
$0.05 / $0.33
329 GPT-5.1 (no reasoning)OpenAI
5.3%
$1.25 / $10
329 gpt-oss-20b (low reasoning)OpenAI
5.3%
$0.03 / $0.15
334 Qwen3.5-0.8B (no reasoning)Alibaba
5.1%
–
334 Gemini 2.5 Flash-Lite Preview (09-2025, no reasoning)Google
5.1%
–
334 GPT-5 mini (minimal reasoning)OpenAI
5.1%
$0.25 / $2
334 Grok 4.1 Fast (non-reasoning, no reasoning)xAI
5.1%
–
338 Qwen3.5-2B (no reasoning)Alibaba
5.0%
–
338 GPT-4.1 miniOpenAI
5.0%
$0.40 / $1.60
338 GLM-4.7-Flash (no reasoning)Z.ai
5.0%
$0.06 / $0.40
341 Llama 4 MaverickMeta
4.9%
$0.27 / $0.85
342 Gemma 4 E2BGoogle
4.8%
–
342 Gemma 4 E4B (no reasoning)Google
4.8%
–
344 DeepSeek-V3-0324DeepSeek
4.7%
$0.25 / $1
344 Gemini 2.5 Flash (no reasoning)Google
4.7%
$0.30 / $2.50
344 Gemma 4 E2B (no reasoning)Google
4.7%
–
344 Mistral Medium 3.1Mistral AI
4.7%
$0.40 / $2
348 Qwen3-30B-A3B (no reasoning)Alibaba
4.6%
$0.12 / $0.50
348 Qwen3-Omni-30B-A3B-InstructAlibaba
4.6%
–
348 Qwen3-VL-4B-ThinkingAlibaba
4.6%
–
348 Nova Micro 1.0Amazon
4.6%
$0.035 / $0.14
348 Ministral 3 14BMistral AI
4.6%
$0.20 / $0.20
348 Nemotron 3 Nano 30B A3B (no reasoning)NVIDIA
4.6%
$0.05 / $0.20
354 Qwen3-14B (reasoning on)Alibaba
4.5%
$0.12 / $0.24
354 Qwen3-4B-Instruct-2507Alibaba
4.5%
–
354 Qwen3-Coder-480B-A35BAlibaba
4.5%
$0.35 / $1.50
354 Llama 3.1 70B InstructMeta
4.5%
$0.40 / $0.40
354 Grok 4 Fast (non-reasoning, no reasoning)xAI
4.5%
–
359 Gemma 3 27BGoogle
4.4%
$0.12 / $0.20
359 Reka Flash 3Reka AI
4.4%
$0.10 / $0.20
361 Nova Lite 1.0Amazon
4.3%
$0.06 / $0.24
361 Claude Sonnet 4 (no reasoning)Anthropic
4.3%
$3 / $15
361 Ministral 3 8BMistral AI
4.3%
$0.15 / $0.15
361 Mistral Small 3.1 24BMistral AI
4.3%
$0.35 / $0.56
361 Mistral Small 3.2 24BMistral AI
4.3%
$0.094 / $0.25
366 Qwen3-235B-A22B (no reasoning)Alibaba
4.2%
$0.46 / $1.82
366 Claude Haiku 4.5 (no reasoning)Anthropic
4.2%
$1 / $5
366 Gemma 3 12BGoogle
4.2%
$0.05 / $0.15
366 Mistral Large 3Mistral AI
4.2%
$0.50 / $1.50
366 GPT-4.1OpenAI
4.2%
$2 / $8
366 GPT-4o mini (2024-07-18)OpenAI
4.2%
$0.15 / $0.60
372 Qwen3-14B (no reasoning)Alibaba
4.1%
$0.12 / $0.24
372 Qwen3-32B (no reasoning)Alibaba
4.1%
$0.14 / $0.40
372 Mistral Medium 3Mistral AI
4.1%
$0.40 / $2
372 GPT-5.4 nano (no reasoning)OpenAI
4.1%
$0.20 / $1.25
376 Nova 2 Lite (low reasoning)Amazon
4.0%
$0.30 / $2.50
376 Command ACohere
4.0%
$2.50 / $10
376 Mixtral 8x22B InstructMistral AI
4.0%
$2 / $6
376 GPT-5 nano (minimal reasoning)OpenAI
4.0%
$0.05 / $0.40
380 Qwen3-8B (reasoning on)Alibaba
3.9%
$0.12 / $0.46
381 Qwen3-Coder-30B-A3B-InstructAlibaba
3.8%
$0.07 / $0.28
381 Qwen3-VL-8B-ThinkingAlibaba
3.8%
$0.18 / $2.10
381 Gemma 4 E4BGoogle
3.8%
–
381 Llama 4 ScoutMeta
3.8%
$0.18 / $0.59
381 Phi-4Microsoft
3.8%
$0.07 / $0.14
381 Devstral MediumMistral AI
3.8%
–
381 Devstral Small 1.1Mistral AI
3.8%
–
381 Mistral Small 3Mistral AI
3.8%
$0.05 / $0.08
381 Mistral Small 4 (no reasoning)Mistral AI
3.8%
$0.15 / $0.60
381 GPT-4.1 nanoOpenAI
3.8%
$0.10 / $0.40
391 Gemini 2.5 Flash-Lite (no reasoning)Google
3.7%
$0.10 / $0.40
391 GLM-4.6V (no reasoning)Z.ai
3.7%
$0.30 / $0.90
393 Qwen2.5-72B-InstructAlibaba
3.6%
$0.36 / $0.40
393 Qwen3-VL-4B-InstructAlibaba
3.6%
–
393 Gemma 3 270MGoogle
3.6%
–
393 Llama 3.3 70B InstructMeta
3.6%
$0.59 / $0.79
393 Devstral 2Mistral AI
3.6%
$0.40 / $2
398 Qwen2.5-Coder-32B-InstructAlibaba
3.5%
$0.66 / $1
398 Devstral Small 2Mistral AI
3.5%
–
398 GLM-4.5V (no reasoning)Z.ai
3.5%
$0.60 / $1.80
401 Nova Pro 1.0Amazon
3.2%
$0.80 / $3.20
402 Nova 2 Lite (no reasoning)Amazon
2.9%
$0.30 / $2.50
402 DeepSeek-V3DeepSeek
2.9%
$0.26 / $1.03
402 Mistral Large 2 (2407)Mistral AI
2.9%
$2 / $6
405 Qwen3-VL-8B-InstructAlibaba
2.7%
$0.12 / $0.46
406 Qwen3.5-2B (reasoning on)Alibaba
2.6%
–
407 Kimi Linear 48B A3B InstructMoonshot AI
2.5%
–
408 GPT-4o (2024-08-06)OpenAI
2.3%
$2.50 / $10
409 Qwen3-8B (no reasoning)Alibaba
1.9%
$0.12 / $0.46
410 GPT-4o (2024-05-13)OpenAI
1.8%
$5 / $15
411 Qwen3.5-0.8B (reasoning on)Alibaba
1.1%
–

Swipe the table sideways for more columns.

Ranks follow the score as shown, so equal numbers share a rank. Results as published by Artificial Analysis; we do not re-run them.

Progress over time

Humanity's Last Exam, plotted at each model's release date, higher is better. Each model counts once, at its best setting. Scores are today's, not launch day's. Hover or tap a mark for its model.

Best score so far

Every model at its release date. The line moves each time a new release beats every earlier one, and each new best is named where there is room.

-20%0%20%40%60%80% Jul 20242025Jul2026Jul GPT-4o mini (2024-07-18)Llama 3.1 8B Instructo1DeepSeek-R1o3-miniGemini 2.5 ProGrok 4GPT-5Gemini 3 Pro PreviewGPT-5.3-CodexGemini 3.1 Pro PreviewClaude Opus 4.8Claude Fable 5Claude Fable 5.1Claude Opus 5.5

Best score by lab

Each lab's best so far, stepping whenever one of its releases beats its own earlier models. The 6 labs with the best scores today; 19 others not drawn.

AnthropicGoogleOpenAIMetaXiaomiMiniMax

-20%0%20%40%60%80% Jul 20242025Jul2026Jul AnthropicGoogleOpenAIMetaXiaomiMiniMax

What it measures

Share of expert-written exam questions answered correctly, as run by Artificial Analysis with its own harness and prompts.

What it does not measure

Not everyday work; academic questions at the edge of expertise.

Source

Artificial Analysis · Methodology

285 results from Artificial Analysis not ranked here · show why

We rank a result only when we can tie it to a specific model you can use. These are left out:

  • Not on sale through the API providers we track: 270
  • A different snapshot or variant from the model we list: 14
  • An unusual combination of settings: 1

Data sourced from Artificial Analysis. Licence: Artificial Analysis commercial data licence.