SShogo Bench

LMCA

Language-model capability assessment across domains.

Agent Index weight
0%
shown for context only
Models tested
140
from the benchmark’s leaderboard and reports
Normalised for the index
0 → 0, 0.85 → 100
Official site

Results

#ModelScoreNormalisedHarnessCost/taskDate
1Claude Fable 5.2 Supported
Anthropic
70.6%source——
2Claude Sonnet 5.5 Estimated
Anthropic
68.3%source——
3Claude Opus 5.5 Supported
Anthropic
68.2%source——
4Claude Fable 5.1 Supported
Anthropic
65.5%source——
5Claude Opus 5 Supported
Anthropic
64.5%source——
6GPT-6 Astra Supported
OpenAI
64.4%source——
7Gemini 4 Argon Supported
Google
63.7%source——
8GPT-6.1 Sol Estimated
OpenAI
62.8%source——
9Claude Fable 5 Supported
Anthropic
61.2%source——
10GPT-5.6 Sol Supported
OpenAI
59.2%source——
11GPT-6 Sol Estimated
OpenAI
59.1%source——
12Kimi K3.1 Pro
Moonshot AI
57.6%source——
13Claude Opus 4.8 Supported
Anthropic
57.5%source——
14Muse Wonder Estimated
Meta
56.8%source——
15Claude Opus 4.6 Supported
Anthropic
55.8%source——
16GLM-5.3 Supported
Z.ai (Zhipu)
55.5%source——
17GPT-5.6 Terra Estimated
OpenAI
54.9%source——
18GPT-5.5 Supported
OpenAI
54.3%source——
19GPT-5.5 Pro
OpenAI
53.9%source——
20Muse Spark 1.3 Estimated
Meta
53.9%source——
21Gemini 3.1 Pro Preview Supported
Google
53.8%source——
22Gemini 3.8 Flash Supported
Google
52.9%source——
23Kimi K3 Supported
Moonshot AI
52.7%source——
24Claude Opus 4.7 Supported
Anthropic
52.2%source——
25GPT-5.4 Supported
OpenAI
52.0%source——
26Claude Haiku 5.5 Estimated
Anthropic
51.1%source——
27Gemini 3.7 Flash Estimated
Google
50.4%source——
28Claude Sonnet 5 Supported
Anthropic
50.0%source——
29Muse Spark 1.1 Estimated
Meta
49.9%source——
30Grok 4.7 Estimated
xAI
49.4%source——
31GPT-5.6 Luna Estimated
OpenAI
48.5%source——
32Grok 4.6 Supported
xAI
48.5%source——
33Muse Spark 1.2 Estimated
Meta
48.4%source——
34Gemini 3.5 Flash Supported
Google
47.1%source——
35DeepSeek V4.1 Flash Estimated
DeepSeek
47.0%source——
36Claude Sonnet 4.6 Supported
Anthropic
46.5%source——
37Qwen3.8 Max Preview Estimated
Alibaba (Qwen)
46.2%source——
38GLM-5.2 Supported
Z.ai (Zhipu)
45.8%source——
39DeepSeek V4 Pro 0813
DeepSeek
45.5%source——
40Grok 4.5 Estimated
xAI
45.2%source——
41Gemini 3.6 Flash Estimated
Google
44.9%source——
42GPT-6 Luna
OpenAI
44.5%source——
43Claude Opus 4.5 (latest) Supported
Anthropic
44.5%source——
44Qwen3.7 Max
Alibaba (Qwen)
44.0%source——
45GPT-5.2 Supported
OpenAI
43.9%source——
46GPT-5.1 Supported
OpenAI
43.9%source——
47Gemini 3 Flash Preview Supported
Google
43.1%source——
48Qwen3.6 Max Preview Estimated
Alibaba (Qwen)
42.5%source——
49DeepSeek V4 Flash 0731
DeepSeek
41.7%source——
50Qwen3.8 27B
Alibaba (Qwen)
41.4%source——
51DeepSeek V4 Pro Estimated
DeepSeek
41.2%source——
52GPT-5.4 mini
OpenAI
40.8%source——
53GPT-5 Supported
OpenAI
40.0%source——
54o3 Supported
OpenAI
39.7%source——
55Gemma 4 31B IT
Google
39.3%source——
56Claude Sonnet 4.5 (latest) Supported
Anthropic
38.8%source——
57Grok 4.20 (Non-Reasoning) Estimated
xAI
38.7%source——
58o3-pro
OpenAI
38.5%source——
59Grok 4.3 Beta
xAI
38.3%source——
60Qwen3.5 397B-A17B
Alibaba (Qwen)
37.9%source——
61Gemini 3.5 Flash Lite
Google
37.6%source——
62Inkling Estimated
thinkingmachines
37.6%source——
63Qwen3.7 Plus
Alibaba (Qwen)
37.6%source——
64Claude-Opus-4 Supported
Anthropic
37.4%source——
65Kimi K2.6 Estimated
Moonshot AI
37.3%source——
66Claude Opus 4.1 (latest) Supported
Anthropic
37.1%source——
67Nemotron 3 Ultra
NVIDIA
36.9%source——
68GPT-5.4 nano
OpenAI
36.9%source——
69Qwen3.5 Plus Estimated
Alibaba (Qwen)
36.4%source——
70DeepSeek V4 Flash
DeepSeek
35.9%source——
71Mistral Large 4
Mistral
35.8%source——
72Gemini 3.1 Flash Lite
Google
35.0%source——
73Gemini 2.5 Pro Supported
Google
34.8%source——
74Qwen3.6 27B
Alibaba (Qwen)
34.5%source——
75GPT-5 Mini Estimated
OpenAI
34.2%source——
76Qwen3.5 27B
Alibaba (Qwen)
34.0%source——
77MiniMax-M3 Estimated
MiniMax
33.7%source——
78Qwen3.6 Plus Estimated
Alibaba (Qwen)
33.1%source——
79Qwen3.5 122B-A10B
Alibaba (Qwen)
32.2%source——
80Qwen3.6 Flash
Alibaba (Qwen)
31.0%source——
81Claude Haiku 4.5 Estimated
Anthropic
30.9%source——
82Qwen3.6 35B-A3B
Alibaba (Qwen)
29.7%source——
83Gemma 4 26B A4B
Google
29.7%source——
84MiMo-V2.5-Pro
Xiaomi
29.5%source——
85Qwen3.5 35B-A3B
Alibaba (Qwen)
29.5%source——
86Qwen3 235B A22B Thinking 2507
Alibaba (Qwen)
29.3%source——
87Qwen3.5 Flash
Alibaba (Qwen)
29.1%source——
88DeepSeek V3.2 Supported
DeepSeek
29.1%source——
89Claude Sonnet 4 Supported
Anthropic
29.0%source——
90DeepSeek V3.1 Terminus
DeepSeek
28.6%source——
91Qwen3 Max Estimated
Alibaba (Qwen)
28.3%source——
92Gemini 2.5 Flash Estimated
Google
27.5%source——
93o4-mini Estimated
OpenAI
26.5%source——
94Mistral Medium 3.5
Mistral
26.1%source——
95GPT-4.1
OpenAI
25.6%source——
96Qwen3 235B A22B
Alibaba (Qwen)
25.0%source——
97Qwen3.5 9B
Alibaba (Qwen)
24.5%source——
98DeepSeek V3.1
DeepSeek
24.3%source——
99Qwen Plus
Alibaba (Qwen)
24.0%source——
100Qwen3 235B A22B Instruct 2507
Alibaba (Qwen)
23.5%source——
101Qwen3 30B A3B Instruct 2507
Alibaba (Qwen)
22.4%source——
102o1 Estimated
OpenAI
22.3%source——
103GPT OSS 120B Supported
OpenAI
22.1%source——
104GPT-4.1 mini
OpenAI
21.1%source——
105Mistral Small 4
Mistral
20.6%source——
106Qwen3-30B-A3B-Thinking (Jul 2025)
Alibaba (Qwen)
19.5%source——
107Mistral Medium 3.1
Mistral
19.1%source——
108o3-mini Estimated
OpenAI
19.0%source——
109Qwen3 14B
Alibaba (Qwen)
18.2%source——
110Gemini 2.5 Flash-Lite
Google
18.1%source——
111Llama 3.3 70B
Meta
17.5%source——
112Qwen3 32B
Alibaba (Qwen)
17.3%source——
113GPT-4
OpenAI
17.1%source——
114Claude 3 Opus Estimated
Anthropic
17.0%source——
115Mistral Large 3
Mistral
16.7%source——
116GPT-4o (2024-08-06) Supported
OpenAI
16.6%source——
117Llama 4 Maverick
Meta
15.9%source——
118Qwen3 30B A3B
Alibaba (Qwen)
15.8%source——
119DeepSeek V3 0324
DeepSeek
15.5%source——
120Llama 3.1-70B
Meta
14.8%source——
121GPT OSS 20B
OpenAI
14.5%source——
122Qwen2.5-72B Estimated
Alibaba (Qwen)
13.4%source——
123Gemma 3 27B
Google
12.3%source——
124Llama 4 Scout
Meta
12.0%source——
125GPT-4o mini
OpenAI
10.4%source——
126Cohere Command A
Cohere
10.3%source——
127GPT-4 Turbo
OpenAI
9.8%source——
128GPT-3.5-turbo
OpenAI
9.7%source——
129c4ai-command-r-08-2024
Unknown
9.2%source——
130Claude-Haiku-3
Anthropic
8.8%source——
131Qwen3 8B
Alibaba (Qwen)
8.8%source——
132GPT-5 Nano
OpenAI
7.9%source——
133Gemma 2 27B
Google
7.1%source——
134Qwen2.5-7B
Alibaba (Qwen)
6.4%source——
135Ministral 3B
Mistral
5.5%source——
136GPT-4.1 nano
OpenAI
5.5%source——
137Llama 3.1-8B
Meta
5.4%source——
138Command R+
Cohere
5.0%source——
139Gemma 3 12B
Google
4.5%source——
140Gemma 3 4B
Google
2.8%source——