Benchmark Explorer

Explore how models perform on various benchmarks

Benchmarks

🔍

Blended Price (USD/1M Tokens)

Calculates a weighted average price per million tokens, assuming a 3:1 ratio of input to output tokens, using the formula: (3 * Input Price + Output Price) / 4.
Source:

Model Performance

#1
Claude 3 Opus
30.0
#2
Claude 4.1 Opus
30.0
#3
Claude 4.1 Opus (Thinking)
30.0
#4
Claude 4.0 Opus (Thinking)
30.0
#5
Claude 4.0 Opus
30.0
#6
OpenAI o1
26.2
#7
Claude Opus 4.6 (Thinking)
10.0
#8
Claude Opus 4.5 (Thinking)
10.0
#9
Claude Opus 4.6
10.0
#10
Claude Opus 4.5
10.0
#11
GPT-4o
7.5
#12
Claude 3.5 Sonnet
6.0
#13
Claude 3.7 Sonnet
6.0
#14
Claude Sonnet 4.6
6.0
#15
Claude Sonnet 4.5 (Thinking)
6.0
#16
Claude 3.7 Sonnet (Thinking)
6.0
#17
Claude 4.0 Sonnet (Thinking)
6.0
#18
Cohere Command R+
6.0
#19
Claude Sonnet 4.5
6.0
#20
Claude 4.0 Sonnet
6.0
#21
Claude 3 Sonnet
6.0
#22
Grok 3
6.0
#23
Grok 4 (Thinking)
6.0
#24
Grok 4
6.0
#25
Mistral Large
6.0
#26
GPT-5.4
5.6
#27
GPT-5.2 Codex
4.8
#28
GPT-5.2
4.8
#29
GPT-5.3 Codex
4.8
#30
Gemini 3.1 Pro Preview
4.5
#31
Cohere Command A
4.4
#32
Llama 3.1 405B
3.7
#33
GPT-4.1
3.5
#34
OpenAI o3 (High Effort)
3.5
#35
OpenAI o3 (Medium Effort)
3.5
#36
GPT-5.1 Codex Max
3.4
#37
Gemini 2.5 Pro (Thinking)
3.4
#38
GPT-5 (Thinking)
3.4
#39
Gemini 2.5 Pro (Jun 2025)
3.4
#40
GPT-5
3.4
#41
Mistral Large 2
3.0
#42
Grok 4.20
3.0
#43
Grok 4.20 (Reasoning)
3.0
#44
Qwen 3 (Thinking)
2.6
#45
Qwen 3 Max Preview
2.4
#46
Qwen 3 Max (Thinking)
2.4
#47
DeepSeek R1
2.4
#48
Claude Haiku 4.5
2.0
#49
Claude Haiku 4.5 (Thinking)
2.0
#50
OpenAI o3 Mini (High Effort)
1.9
#51
OpenAI o4 Mini (High Effort)
1.9
#52
OpenAI o3 Mini (Medium Effort)
1.9
#53
OpenAI o4 Mini (Medium Effort)
1.9
#54
GPT-5.4 Mini
1.7
#55
Claude 3.5 Haiku
1.6
#56
GLM-5
1.6
#57
DeepSeek V3 (Mar 2025)
1.2
#58
Qwen 3
1.2
#59
Kimi K2.5
1.2
#60
Gemini 3.0 Flash
1.1
#61
Kimi K2
1.0
#62
GLM-4.7
0.9
#63
Gemini 2.5 Flash
0.8
#64
Gemini 2.5 Flash (Thinking)
0.8
#65
Mistral Medium 3.1
0.8
#66
Mistral Large 3
0.8
#67
GPT-3.5 Turbo
0.8
#68
GPT-5 Mini (Thinking)
0.7
#69
GPT-5 Mini
0.7
#70
Llama 3.3 70B
0.6
#71
DeepSeek V3
0.6
#72
Gemini 3.1 Flash Lite Preview
0.6
#73
MiniMax M2.1
0.5
#74
MiniMax M2.5
0.5
#75
MiniMax M2.7
0.5
#76
Llama 4 Maverick 17B
0.5
#77
GPT-5.4 Nano
0.5
#78
Grok 3 Mini
0.3
#79
DeepSeek V3.2
0.3
#80
DeepSeek V3.2 (Thinking)
0.3
#81
GPT-4o Mini
0.3
#82
Gemini 2.0 Flash
0.3
#83
GPT OSS 120B
0.3
#84
Phi-4
0.2
#85
MiMo V2 Flash
0.1
#86
GPT-5 Nano
0.1
#87
GPT-5 Nano (Thinking)
0.1
#88
Llama 2 7B
0.1
#89
Devstral 2
0.0
#90
GPT-4.5
0.0
#91
Devstral Small 2
0.0
#92
Gemini 2.0 Pro
0.0
#93
Gemini 1.5 Pro
0.0
#94
Grok 3 (Thinking)
0.0
#95
Gemini 2.0 Flash Thinking (Jan 2025)
0.0
#96
Gemini 2.0 Flash (Thinking)
0.0
#97
Gemini 2.5 Pro
0.0
#98
Magistral Medium 3.1
0.0
#99
Llama 2 13B
0.0
#100
Llama 2 70B
0.0
#101
OpenAI o1 Mini
0.0