context.vn
Menu

bullshit Bench V2

96 models evaluated

#ModelProviderTypeScore
1Claude Opus 4.8AnthropicProprietary95%
2Qwen3.8 MaxAlibabaOpen weights95%
4Claude Sonnet 4.6Anthropic · Closed91%
5Claude Opus 4.5Anthropic · Closed90%
7Claude Opus 4.6Anthropic · Closed87%
9Claude Opus 4.7Anthropic · Closed83%
10Claude Sonnet 5Anthropic · Closed80%
11Claude Sonnet 4.5Anthropic · Closed79%
13Qwen3.8-27BAlibaba · Open weight79%
15Qwen3.5 397B (Reasoning)Alibaba · Open weight78%
16Claude Fable 5.1Anthropic · Closed77%
17Claude Haiku 4.5Anthropic · Closed77%
18Kimi K3Moonshot AI · Closed74%
21Claude Opus 5Anthropic · Closed73%
23GLM-5.3Z.AI · Open weight72%
24Qwen3.7 MaxAlibaba · Closed72%
25Qwen3.6 PlusAlibaba · Closed72%
28GPT-6 AstraOpenAI · Closed69%
29Qwen3.5 397BAlibaba · Open weight69%
30Grok 4.20 Multi-agentxAI · Closed67%
32Gemini 3.5 Flash-LiteGoogle · Closed66%
33Grok 4.6xAI · Closed66%
34Kimi K2.6Moonshot AI · Open weight65%
35stealth/ox-alpha (low)Stealth65%
38MiniMax M3MiniMax · Open weight64%
42MiMo-V2.5-ProXiaomi · Closed62%
45DeepSeek V4.1 FlashDeepSeek · Open weight57%
48Claude Fable 5Anthropic · Closed56%
50Grok 4.20xAI · Closed56%
51Grok 4.5xAI · Closed55%
53GPT-5.6 TerraOpenAI · Closed54%
56Nemotron 3 Super 120B A12BNVIDIA · Open weight54%
57Kimi K2.5Moonshot AI · Open weight52%
58Grok 4.3xAI · Closed50%
60Muse Spark 1.2Meta · Closed50%
65GPT-5.6 SolOpenAI · Closed48%
66GPT-5.4OpenAI · Closed48%
67Gemini 3 ProGoogle · Closed48%
69GPT-5.5OpenAI · Closed47%
74Claude 3.5 SonnetAnthropic · Closed46%
78GPT-5.2-CodexOpenAI · Closed45%
79GPT-5.1OpenAI · Closed45%
81Claude 4.1 OpusAnthropic · Closed43%
86GPT-5.6 LunaOpenAI · Closed41%
87GPT-5.3 InstantOpenAI · Closed40%
89DeepSeek V4 Flash 0731DeepSeek · Closed39%
92Gemini 3.6 FlashGoogle · Closed39%
93GPT-5.2OpenAI · Closed38%
96Gemini 3.1 ProGoogle · Closed37%
99GPT-5.5 ProOpenAI · Closed36%
100Gemini 3 Pro Deep ThinkGoogle · Closed36%
101DeepSeek V4 Pro 0813DeepSeek · Closed35%
102Gemini 3.7 FlashGoogle · Closed35%
103MiMo-V2.5Xiaomi · Closed35%
108GPT-5.4 miniOpenAI · Closed32%
111GPT-5.1-Codex-MaxOpenAI · Closed32%
113Kimi K2.5 (Reasoning)Moonshot AI · Closed31%
116GLM-5-TurboZ.AI · Closed31%
118GLM-5.2Z.AI · Open weight31%
119Claude 4 SonnetAnthropic · Closed30%
122Llama 4 MaverickMeta · Open weight29%
125GLM-5 (Reasoning)Z.AI · Open weight28%
128GPT-5.2 InstantOpenAI · Closed27%
129o3OpenAI · Closed26%
132Gemma 4 31BGoogle · Open weight25%
133GPT-5.3 CodexOpenAI · Closed24%
136GLM-5.1Z.AI · Open weight22%
137Step 3.5 FlashStepFun · Open weight22%
138Laguna S 2.1Poolside · Open weight22%
140Gemma 4 26B A4BGoogle · Open weight21%
145Gemini 2.5 ProGoogle · Closed20%
146GLM-5Z.AI · Open weight20%
147Llama 4 ScoutMeta · Open weight20%
149Gemini 3.5 FlashGoogle · Closed20%
151Grok 4.1 FastxAI · Closed19%
152Gemini 2.5 FlashGoogle · Closed19%
158Trinity-Large-ThinkingArcee AI · Open weight17%
159MiMo-V2-FlashXiaomi · Open weight16%
160Hy3Tencent · Open weight16%
164GPT-5.4 nanoOpenAI · Closed14%
167GPT-4.1OpenAI · Closed14%
169DeepSeek V3.2 (Thinking)DeepSeek · Open weight13%
173GPT-4oOpenAI · Closed12%
174GPT-OSS 120BOpenAI · Open weight12%
176Gemini 3.1 Flash-LiteGoogle · Closed11%
177Seed 1.6ByteDance · Closed11%
180Gemini 3 FlashGoogle · Closed10%
181DeepSeek V3.2DeepSeek · Open weight10%
182Claude 3 HaikuAnthropic · Closed10%
185Kimi K2Moonshot AI · Closed10%
187MiniMax M2.5MiniMax · Closed9%
190GLM-4.5Z.AI · Closed8%
191MiniMax M2.7MiniMax · Open weight8%
192DeepSeek-R1DeepSeek · Open weight8%
193o4-mini (high)OpenAI · Closed8%
214GPT-4o miniOpenAI · Closed2%