File size: 5,726 Bytes
3c1db6c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
[
  {
    "model_name": "gpt-4o",
    "provider": "openai",
    "input_cost_per_1k": 0.0025,
    "output_cost_per_1k": 0.010,
    "context_window": 128000,
    "reasoning_score": 0.92,
    "coding_score": 0.93,
    "math_score": 0.90,
    "instruction_following_score": 0.95,
    "latency_score": 0.70,
    "max_complexity": 1.0,
    "notes": "Flagship multimodal model. Best quality/cost for hard tasks."
  },
  {
    "model_name": "gpt-4o-mini",
    "provider": "openai",
    "input_cost_per_1k": 0.000150,
    "output_cost_per_1k": 0.000600,
    "context_window": 128000,
    "reasoning_score": 0.78,
    "coding_score": 0.78,
    "math_score": 0.72,
    "instruction_following_score": 0.85,
    "latency_score": 0.90,
    "max_complexity": 0.75,
    "notes": "Cost-efficient small model for lightweight tasks."
  },
  {
    "model_name": "gpt-3.5-turbo",
    "provider": "openai",
    "input_cost_per_1k": 0.0005,
    "output_cost_per_1k": 0.0015,
    "context_window": 16385,
    "reasoning_score": 0.62,
    "coding_score": 0.65,
    "math_score": 0.55,
    "instruction_following_score": 0.75,
    "latency_score": 0.92,
    "max_complexity": 0.60,
    "notes": "Legacy fast model. Good for simple chat tasks."
  },
  {
    "model_name": "claude-3-5-haiku-20241022",
    "provider": "anthropic",
    "input_cost_per_1k": 0.00080,
    "output_cost_per_1k": 0.00400,
    "context_window": 200000,
    "reasoning_score": 0.78,
    "coding_score": 0.80,
    "math_score": 0.75,
    "instruction_following_score": 0.85,
    "latency_score": 0.92,
    "max_complexity": 0.75,
    "notes": "Fast, affordable Anthropic model for everyday tasks."
  },
  {
    "model_name": "claude-3-5-sonnet-20241022",
    "provider": "anthropic",
    "input_cost_per_1k": 0.003,
    "output_cost_per_1k": 0.015,
    "context_window": 200000,
    "reasoning_score": 0.93,
    "coding_score": 0.95,
    "math_score": 0.88,
    "instruction_following_score": 0.96,
    "latency_score": 0.75,
    "max_complexity": 1.0,
    "notes": "Top-tier coding and reasoning model from Anthropic."
  },
  {
    "model_name": "claude-3-haiku-20240307",
    "provider": "anthropic",
    "input_cost_per_1k": 0.00025,
    "output_cost_per_1k": 0.00125,
    "context_window": 200000,
    "reasoning_score": 0.65,
    "coding_score": 0.65,
    "math_score": 0.60,
    "instruction_following_score": 0.75,
    "latency_score": 0.95,
    "max_complexity": 0.60,
    "notes": "Cheapest Anthropic model. Good for classification, summarization."
  },
  {
    "model_name": "gemini-1.5-flash",
    "provider": "google",
    "input_cost_per_1k": 0.000075,
    "output_cost_per_1k": 0.000300,
    "context_window": 1000000,
    "reasoning_score": 0.74,
    "coding_score": 0.74,
    "math_score": 0.70,
    "instruction_following_score": 0.78,
    "latency_score": 0.88,
    "max_complexity": 0.72,
    "notes": "Extremely cheap and fast. Long context support."
  },
  {
    "model_name": "gemini-1.5-pro",
    "provider": "google",
    "input_cost_per_1k": 0.00125,
    "output_cost_per_1k": 0.005,
    "context_window": 2000000,
    "reasoning_score": 0.88,
    "coding_score": 0.87,
    "math_score": 0.85,
    "instruction_following_score": 0.90,
    "latency_score": 0.72,
    "max_complexity": 0.95,
    "notes": "Massive context window. Great for long-doc analysis."
  },
  {
    "model_name": "mistral-small-latest",
    "provider": "mistral",
    "input_cost_per_1k": 0.001,
    "output_cost_per_1k": 0.003,
    "context_window": 32000,
    "reasoning_score": 0.68,
    "coding_score": 0.70,
    "math_score": 0.62,
    "instruction_following_score": 0.75,
    "latency_score": 0.88,
    "max_complexity": 0.65,
    "notes": "Cost-effective European model."
  },
  {
    "model_name": "mistral-large-latest",
    "provider": "mistral",
    "input_cost_per_1k": 0.003,
    "output_cost_per_1k": 0.009,
    "context_window": 128000,
    "reasoning_score": 0.85,
    "coding_score": 0.84,
    "math_score": 0.80,
    "instruction_following_score": 0.88,
    "latency_score": 0.75,
    "max_complexity": 0.90,
    "notes": "Strong European flagship model."
  },
  {
    "model_name": "llama3.2:3b",
    "provider": "ollama",
    "input_cost_per_1k": 0.0,
    "output_cost_per_1k": 0.0,
    "context_window": 128000,
    "reasoning_score": 0.50,
    "coding_score": 0.48,
    "math_score": 0.42,
    "instruction_following_score": 0.60,
    "latency_score": 0.95,
    "max_complexity": 0.45,
    "notes": "Free local model. Use for simple/private tasks."
  },
  {
    "model_name": "llama3.1:8b",
    "provider": "ollama",
    "input_cost_per_1k": 0.0,
    "output_cost_per_1k": 0.0,
    "context_window": 128000,
    "reasoning_score": 0.65,
    "coding_score": 0.64,
    "math_score": 0.58,
    "instruction_following_score": 0.72,
    "latency_score": 0.85,
    "max_complexity": 0.62,
    "notes": "Free local model with decent reasoning."
  },
  {
    "model_name": "llama3.1:70b",
    "provider": "ollama",
    "input_cost_per_1k": 0.0,
    "output_cost_per_1k": 0.0,
    "context_window": 128000,
    "reasoning_score": 0.82,
    "coding_score": 0.82,
    "math_score": 0.78,
    "instruction_following_score": 0.85,
    "latency_score": 0.55,
    "max_complexity": 0.85,
    "notes": "Free local large model. Needs beefy hardware."
  },
  {
    "model_name": "deepseek-chat",
    "provider": "deepseek",
    "input_cost_per_1k": 0.00014,
    "output_cost_per_1k": 0.00028,
    "context_window": 64000,
    "reasoning_score": 0.88,
    "coding_score": 0.90,
    "math_score": 0.92,
    "instruction_following_score": 0.85,
    "latency_score": 0.72,
    "max_complexity": 0.92,
    "notes": "Exceptional value model especially strong in math and code."
  }
]