SShogo Bench

Fiction.LiveBench

Deep comprehension of long stories.

Agent Index weight
0%
shown for context only
Models tested
51
from the benchmark’s leaderboard and reports
Normalised for the index
0 → 0, 1 → 100
Official site

Results

#ModelScoreNormalisedHarnessCost/taskDate
1GPT-5 Supported
OpenAI
97.2%source——
2o3-pro
OpenAI
97.2%source——
3Grok 4 Supported
xAI
94.4%source——
4Grok 4 Fast
xAI
94.4%source——
5Gemini 2.5 Pro Supported
Google
91.7%source——
6o3 Supported
OpenAI
88.9%source——
7Kimi K2.5 Supported
Moonshot AI
86.1%source——
8Claude 3.7 Sonnet Supported
Anthropic
83.3%source——
9DeepSeek V3.2 Supported
DeepSeek
83.3%source——
10o1 Estimated
OpenAI
83.3%source——
11Qwq 32B
Alibaba (Qwen)
83.3%source——
12Gemini 2.5 Flash Estimated
Google
77.8%source——
13o4-mini Estimated
OpenAI
77.8%source——
14DeepSeek-R1 Estimated
DeepSeek
75.0%source——
15Qwen3 235B A22B Thinking 2507
Alibaba (Qwen)
75.0%source——
16Qwen3 32B
Alibaba (Qwen)
74.2%source——
17GPT-5 Mini Estimated
OpenAI
69.4%source——
18MiniMax M1
MiniMax
69.4%source——
19Qwen3 235B A22B
Alibaba (Qwen)
67.7%source——
20GPT-4o (2024-08-06) Supported
OpenAI
66.7%source——
21Grok-3 mini
xAI
66.7%source——
22Kimi K2 0905 Estimated
Moonshot AI
66.7%source——
23Qwen2.5-Max
Alibaba (Qwen)
66.7%source——
24Qwen3 Max Estimated
Alibaba (Qwen)
66.7%source——
25GPT-4.1
OpenAI
63.9%source——
26GPT-4.5
OpenAI
63.9%source——
27Qwen3 14B
Alibaba (Qwen)
62.5%source——
28Qwen3 8B
Alibaba (Qwen)
62.1%source——
29Claude-Opus-4 Supported
Anthropic
61.1%source——
30Gemini-2.0-Flash
Google
61.1%source——
31chutes/GLM-4.5-FP8
Unknown
58.3%source——
32Grok 3
xAI
58.3%source——
33Qwen3-Next 80B-A3B Instruct
Alibaba (Qwen)
55.6%source——
34Qwen3 235B A22B Instruct 2507
Alibaba (Qwen)
52.9%source——
35DeepSeek V3.1
DeepSeek
52.8%source——
36Gemini 2.0 Flash Thinking (Jan 2025)
Google
52.8%source——
37DeepSeek V3 0324
DeepSeek
50.0%source——
38o3-mini Estimated
OpenAI
50.0%source——
39Gemini 2.5 Flash-Lite
Google
47.2%source——
40Claude Sonnet 4 Supported
Anthropic
46.9%source——
41Llama 4 Maverick
Meta
46.2%source——
42GPT-4.1 mini
OpenAI
44.4%source——
43GPT-5 Nano
OpenAI
44.4%source——
44GPT OSS 120B Supported
OpenAI
44.4%source——
45Gemini 2.0 Pro
Google
41.7%source——
46Qwen3 30B A3B
Alibaba (Qwen)
40.6%source——
47Llama 4 Scout
Meta
36.0%source——
48Gemma 3 27B
Google
33.3%source——
49Llama 3.3 70B
Meta
33.3%source——
50GPT-4.1 nano
OpenAI
25.0%source——
51Nvidia Nemotron Nano 9B V2
NVIDIA
25.0%source——