mirascope 1.19.0__py3-none-any.whl → 1.20.1__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (90) hide show
  1. mirascope/__init__.py +4 -0
  2. mirascope/beta/openai/realtime/realtime.py +7 -8
  3. mirascope/beta/openai/realtime/tool.py +2 -2
  4. mirascope/core/__init__.py +10 -1
  5. mirascope/core/anthropic/_utils/__init__.py +0 -2
  6. mirascope/core/anthropic/_utils/_convert_message_params.py +1 -7
  7. mirascope/core/anthropic/_utils/_message_param_converter.py +48 -31
  8. mirascope/core/anthropic/call_response.py +7 -9
  9. mirascope/core/anthropic/call_response_chunk.py +10 -0
  10. mirascope/core/anthropic/stream.py +6 -8
  11. mirascope/core/azure/_utils/__init__.py +0 -2
  12. mirascope/core/azure/call_response.py +7 -10
  13. mirascope/core/azure/call_response_chunk.py +6 -1
  14. mirascope/core/azure/stream.py +6 -8
  15. mirascope/core/base/__init__.py +10 -1
  16. mirascope/core/base/_utils/__init__.py +2 -0
  17. mirascope/core/base/_utils/_get_image_dimensions.py +39 -0
  18. mirascope/core/base/call_response.py +36 -6
  19. mirascope/core/base/call_response_chunk.py +15 -1
  20. mirascope/core/base/stream.py +25 -3
  21. mirascope/core/base/types.py +276 -2
  22. mirascope/core/bedrock/_utils/__init__.py +0 -2
  23. mirascope/core/bedrock/call_response.py +7 -10
  24. mirascope/core/bedrock/call_response_chunk.py +6 -0
  25. mirascope/core/bedrock/stream.py +6 -10
  26. mirascope/core/cohere/_utils/__init__.py +0 -2
  27. mirascope/core/cohere/call_response.py +7 -10
  28. mirascope/core/cohere/call_response_chunk.py +6 -0
  29. mirascope/core/cohere/stream.py +5 -8
  30. mirascope/core/costs/__init__.py +5 -0
  31. mirascope/core/{anthropic/_utils/_calculate_cost.py → costs/_anthropic_calculate_cost.py} +45 -14
  32. mirascope/core/{azure/_utils/_calculate_cost.py → costs/_azure_calculate_cost.py} +3 -3
  33. mirascope/core/{bedrock/_utils/_calculate_cost.py → costs/_bedrock_calculate_cost.py} +3 -3
  34. mirascope/core/{cohere/_utils/_calculate_cost.py → costs/_cohere_calculate_cost.py} +12 -8
  35. mirascope/core/{gemini/_utils/_calculate_cost.py → costs/_gemini_calculate_cost.py} +7 -7
  36. mirascope/core/costs/_google_calculate_cost.py +427 -0
  37. mirascope/core/costs/_groq_calculate_cost.py +156 -0
  38. mirascope/core/costs/_litellm_calculate_cost.py +11 -0
  39. mirascope/core/costs/_mistral_calculate_cost.py +64 -0
  40. mirascope/core/costs/_openai_calculate_cost.py +416 -0
  41. mirascope/core/{vertex/_utils/_calculate_cost.py → costs/_vertex_calculate_cost.py} +8 -7
  42. mirascope/core/{xai/_utils/_calculate_cost.py → costs/_xai_calculate_cost.py} +9 -9
  43. mirascope/core/costs/calculate_cost.py +86 -0
  44. mirascope/core/gemini/_utils/__init__.py +0 -2
  45. mirascope/core/gemini/call_response.py +7 -10
  46. mirascope/core/gemini/call_response_chunk.py +6 -1
  47. mirascope/core/gemini/stream.py +5 -8
  48. mirascope/core/google/_utils/__init__.py +0 -2
  49. mirascope/core/google/_utils/_setup_call.py +21 -2
  50. mirascope/core/google/call_response.py +9 -10
  51. mirascope/core/google/call_response_chunk.py +6 -1
  52. mirascope/core/google/stream.py +5 -8
  53. mirascope/core/groq/_utils/__init__.py +0 -2
  54. mirascope/core/groq/call_response.py +22 -10
  55. mirascope/core/groq/call_response_chunk.py +6 -0
  56. mirascope/core/groq/stream.py +5 -8
  57. mirascope/core/litellm/call_response.py +3 -4
  58. mirascope/core/litellm/stream.py +30 -22
  59. mirascope/core/mistral/_utils/__init__.py +0 -2
  60. mirascope/core/mistral/call_response.py +7 -10
  61. mirascope/core/mistral/call_response_chunk.py +6 -0
  62. mirascope/core/mistral/stream.py +5 -8
  63. mirascope/core/openai/_utils/__init__.py +0 -2
  64. mirascope/core/openai/_utils/_convert_message_params.py +4 -4
  65. mirascope/core/openai/call_response.py +30 -10
  66. mirascope/core/openai/call_response_chunk.py +6 -0
  67. mirascope/core/openai/stream.py +5 -8
  68. mirascope/core/vertex/_utils/__init__.py +0 -2
  69. mirascope/core/vertex/call_response.py +5 -10
  70. mirascope/core/vertex/call_response_chunk.py +6 -0
  71. mirascope/core/vertex/stream.py +5 -8
  72. mirascope/core/xai/_utils/__init__.py +1 -2
  73. mirascope/core/xai/call_response.py +0 -11
  74. mirascope/llm/__init__.py +10 -2
  75. mirascope/llm/_protocols.py +8 -28
  76. mirascope/llm/call_response.py +6 -6
  77. mirascope/llm/call_response_chunk.py +12 -3
  78. mirascope/llm/llm_call.py +21 -23
  79. mirascope/llm/llm_override.py +56 -27
  80. mirascope/llm/stream.py +7 -7
  81. mirascope/llm/tool.py +1 -1
  82. mirascope/retries/fallback.py +1 -1
  83. {mirascope-1.19.0.dist-info → mirascope-1.20.1.dist-info}/METADATA +1 -1
  84. {mirascope-1.19.0.dist-info → mirascope-1.20.1.dist-info}/RECORD +86 -82
  85. mirascope/core/google/_utils/_calculate_cost.py +0 -215
  86. mirascope/core/groq/_utils/_calculate_cost.py +0 -69
  87. mirascope/core/mistral/_utils/_calculate_cost.py +0 -48
  88. mirascope/core/openai/_utils/_calculate_cost.py +0 -246
  89. {mirascope-1.19.0.dist-info → mirascope-1.20.1.dist-info}/WHEEL +0 -0
  90. {mirascope-1.19.0.dist-info → mirascope-1.20.1.dist-info}/licenses/LICENSE +0 -0
mirascope/core/google/_utils/_calculate_cost.py DELETED
@@ -1,215 +0,0 @@
1
- """Calculate the cost of a Gemini API call."""
2
-
3
-
4
- def calculate_cost(
5
- input_tokens: int | float | None,
6
- cached_tokens: int | float | None,
7
- output_tokens: int | float | None,
8
- model: str,
9
- ) -> float | None:
10
- """Calculate the cost of a Google API call.
11
-
12
- https://ai.google.dev/pricing
13
-
14
- Pricing (per 1M tokens):
15
-
16
- Model Input (<128K) Output (<128K) Input (>128K) Output (>128K) Cached
17
- gemini-2.0-pro $1.25 $5.00 $2.50 $10.00 $0.625
18
- gemini-2.0-pro-preview-1206 $1.25 $5.00 $2.50 $10.00 $0.625
19
- gemini-2.0-flash $0.10 $0.40 $0.10 $0.40 $0.0375
20
- gemini-2.0-flash-latest $0.10 $0.40 $0.10 $0.40 $0.0375
21
- gemini-2.0-flash-001 $0.10 $0.40 $0.10 $0.40 $0.0375
22
- gemini-2.0-flash-lite $0.075 $0.30 $0.075 $0.30 $0.0375
23
- gemini-2.0-flash-lite-preview-02-05 $0.075 $0.30 $0.075 $0.30 $0.0375
24
- gemini-1.5-pro $1.25 $5.00 $2.50 $10.00 $0.625
25
- gemini-1.5-pro-latest $1.25 $5.00 $2.50 $10.00 $0.625
26
- gemini-1.5-pro-001 $1.25 $5.00 $2.50 $10.00 $0.625
27
- gemini-1.5-pro-002 $1.25 $5.00 $2.50 $10.00 $0.625
28
- gemini-1.5-flash $0.075 $0.30 $0.15 $0.60 $0.0375
29
- gemini-1.5-flash-latest $0.075 $0.30 $0.15 $0.60 $0.0375
30
- gemini-1.5-flash-001 $0.075 $0.30 $0.15 $0.60 $0.0375
31
- gemini-1.5-flash-002 $0.075 $0.30 $0.15 $0.60 $0.0375
32
- gemini-1.5-flash-8b $0.0375 $0.15 $0.075 $0.30 $0.025
33
- gemini-1.5-flash-8b-latest $0.0375 $0.15 $0.075 $0.30 $0.025
34
- gemini-1.5-flash-8b-001 $0.0375 $0.15 $0.075 $0.30 $0.025
35
- gemini-1.5-flash-8b-002 $0.0375 $0.15 $0.075 $0.30 $0.025
36
- gemini-1.0-pro $0.50 $1.50 $0.50 $1.50 $0.00
37
-
38
- Args:
39
- input_tokens: Number of input tokens
40
- cached_tokens: Number of cached tokens
41
- output_tokens: Number of output tokens
42
- model: Model name to use for pricing calculation
43
-
44
- Returns:
45
- Total cost in USD or None if invalid input
46
- """
47
- pricing = {
48
- "gemini-2.0-pro": {
49
- "prompt_short": 0.000_001_25,
50
- "completion_short": 0.000_005,
51
- "prompt_long": 0.000_002_5,
52
- "completion_long": 0.000_01,
53
- "cached": 0.000_000_625,
54
- },
55
- "gemini-2.0-pro-preview-1206": {
56
- "prompt_short": 0.000_001_25,
57
- "completion_short": 0.000_005,
58
- "prompt_long": 0.000_002_5,
59
- "completion_long": 0.000_01,
60
- "cached": 0.000_000_625,
61
- },
62
- "gemini-2.0-flash": {
63
- "prompt_short": 0.000_000_10,
64
- "completion_short": 0.000_000_40,
65
- "prompt_long": 0.000_000_10,
66
- "completion_long": 0.000_000_40,
67
- "cached": 0.000_000_037_5,
68
- },
69
- "gemini-2.0-flash-latest": {
70
- "prompt_short": 0.000_000_10,
71
- "completion_short": 0.000_000_40,
72
- "prompt_long": 0.000_000_10,
73
- "completion_long": 0.000_000_40,
74
- "cached": 0.000_000_037_5,
75
- },
76
- "gemini-2.0-flash-001": {
77
- "prompt_short": 0.000_000_10,
78
- "completion_short": 0.000_000_40,
79
- "prompt_long": 0.000_000_10,
80
- "completion_long": 0.000_000_40,
81
- "cached": 0.000_000_037_5,
82
- },
83
- "gemini-2.0-flash-lite": {
84
- "prompt_short": 0.000_000_075,
85
- "completion_short": 0.000_000_30,
86
- "prompt_long": 0.000_000_075,
87
- "completion_long": 0.000_000_30,
88
- "cached": 0.000_000_037_5,
89
- },
90
- "gemini-2.0-flash-lite-preview-02-05": {
91
- "prompt_short": 0.000_000_075,
92
- "completion_short": 0.000_000_30,
93
- "prompt_long": 0.000_000_075,
94
- "completion_long": 0.000_000_30,
95
- "cached": 0.000_000_037_5,
96
- },
97
- "gemini-1.5-pro": {
98
- "prompt_short": 0.000_001_25,
99
- "completion_short": 0.000_005,
100
- "prompt_long": 0.000_002_5,
101
- "completion_long": 0.000_01,
102
- "cached": 0.000_000_625,
103
- },
104
- "gemini-1.5-pro-latest": {
105
- "prompt_short": 0.000_001_25,
106
- "completion_short": 0.000_005,
107
- "prompt_long": 0.000_002_5,
108
- "completion_long": 0.000_01,
109
- "cached": 0.000_000_625,
110
- },
111
- "gemini-1.5-pro-001": {
112
- "prompt_short": 0.000_001_25,
113
- "completion_short": 0.000_005,
114
- "prompt_long": 0.000_002_5,
115
- "completion_long": 0.000_01,
116
- "cached": 0.000_000_625,
117
- },
118
- "gemini-1.5-pro-002": {
119
- "prompt_short": 0.000_001_25,
120
- "completion_short": 0.000_005,
121
- "prompt_long": 0.000_002_5,
122
- "completion_long": 0.000_01,
123
- "cached": 0.000_000_625,
124
- },
125
- "gemini-1.5-flash": {
126
- "prompt_short": 0.000_000_075,
127
- "completion_short": 0.000_000_30,
128
- "prompt_long": 0.000_000_15,
129
- "completion_long": 0.000_000_60,
130
- "cached": 0.000_000_037_5,
131
- },
132
- "gemini-1.5-flash-latest": {
133
- "prompt_short": 0.000_000_075,
134
- "completion_short": 0.000_000_30,
135
- "prompt_long": 0.000_000_15,
136
- "completion_long": 0.000_000_60,
137
- "cached": 0.000_000_037_5,
138
- },
139
- "gemini-1.5-flash-001": {
140
- "prompt_short": 0.000_000_075,
141
- "completion_short": 0.000_000_30,
142
- "prompt_long": 0.000_000_15,
143
- "completion_long": 0.000_000_60,
144
- "cached": 0.000_000_037_5,
145
- },
146
- "gemini-1.5-flash-002": {
147
- "prompt_short": 0.000_000_075,
148
- "completion_short": 0.000_000_30,
149
- "prompt_long": 0.000_000_15,
150
- "completion_long": 0.000_000_60,
151
- "cached": 0.000_000_037_5,
152
- },
153
- "gemini-1.5-flash-8b": {
154
- "prompt_short": 0.000_000_037_5,
155
- "completion_short": 0.000_000_15,
156
- "prompt_long": 0.000_000_075,
157
- "completion_long": 0.000_000_30,
158
- "cached": 0.000_000_025,
159
- },
160
- "gemini-1.5-flash-8b-latest": {
161
- "prompt_short": 0.000_000_037_5,
162
- "completion_short": 0.000_000_15,
163
- "prompt_long": 0.000_000_075,
164
- "completion_long": 0.000_000_30,
165
- "cached": 0.000_000_025,
166
- },
167
- "gemini-1.5-flash-8b-001": {
168
- "prompt_short": 0.000_000_037_5,
169
- "completion_short": 0.000_000_15,
170
- "prompt_long": 0.000_000_075,
171
- "completion_long": 0.000_000_30,
172
- "cached": 0.000_000_025,
173
- },
174
- "gemini-1.5-flash-8b-002": {
175
- "prompt_short": 0.000_000_037_5,
176
- "completion_short": 0.000_000_15,
177
- "prompt_long": 0.000_000_075,
178
- "completion_long": 0.000_000_30,
179
- "cached": 0.000_000_025,
180
- },
181
- "gemini-1.0-pro": {
182
- "prompt_short": 0.000_000_5,
183
- "completion_short": 0.000_001_5,
184
- "prompt_long": 0.000_000_5,
185
- "completion_long": 0.000_001_5,
186
- "cached": 0.000_000,
187
- },
188
- }
189
-
190
- if input_tokens is None or output_tokens is None:
191
- return None
192
-
193
- if cached_tokens is None:
194
- cached_tokens = 0
195
-
196
- try:
197
- model_pricing = pricing[model]
198
- except KeyError:
199
- return None
200
-
201
- # Determine if we're using long context pricing
202
- use_long_context = input_tokens > 128_000
203
-
204
- prompt_price = model_pricing["prompt_long" if use_long_context else "prompt_short"]
205
- cached_price = model_pricing["cached"]
206
- completion_price = model_pricing[
207
- "completion_long" if use_long_context else "completion_short"
208
- ]
209
-
210
- prompt_cost = input_tokens * prompt_price
211
- cached_cost = cached_tokens * cached_price
212
- completion_cost = output_tokens * completion_price
213
- total_cost = prompt_cost + cached_cost + completion_cost
214
-
215
- return total_cost
mirascope/core/groq/_utils/_calculate_cost.py DELETED
@@ -1,69 +0,0 @@
1
- """Calculate the cost of a completion using the Groq API."""
2
-
3
-
4
- def calculate_cost(
5
- input_tokens: int | float | None,
6
- cached_tokens: int | float | None,
7
- output_tokens: int | float | None,
8
- model: str = "mixtral-8x7b-32768",
9
- ) -> float | None:
10
- """Calculate the cost of a completion using the Groq API.
11
-
12
- https://wow.groq.com/
13
-
14
- Model Input Cached Output
15
- llama-3.1-405b-reasoning N/A N/A
16
- llama-3.1-70b-versatile N/A N/A
17
- llama-3.1-8b-instant N/A N/A
18
- llama3-groq-70b-8192-tool-use-preview $0.89 / 1M tokens $0.89 / 1M tokens
19
- llama3-groq-8b-8192-tool-use-preview $0.19 / 1M tokens $0.19 / 1M tokens
20
- llama3-70b-8192 $0.59 / 1M tokens $0.79 / 1M tokens
21
- llama3-8b-8192 $0.05 / 1M tokens $0.08 / 1M tokens
22
- mixtral-8x7b-32768 $0.27 / 1M tokens $0.27 / 1M tokens
23
- gemma-7b-it $0.07 / 1M tokens $0.07 / 1M tokens
24
- gemma2-9b-it $0.20 / 1M tokens $0.20 / 1M tokens
25
- """
26
- pricing = {
27
- "llama3-groq-70b-8192-tool-use-preview": {
28
- "prompt": 0.000_000_89,
29
- "completion": 0.000_000_89,
30
- },
31
- "llama3-groq-8b-8192-tool-use-preview": {
32
- "prompt": 0.000_000_19,
33
- "completion": 0.000_000_19,
34
- },
35
- "llama3-70b-8192": {
36
- "prompt": 0.000_000_59,
37
- "completion": 0.000_000_79,
38
- },
39
- "llama3-8b-8192": {
40
- "prompt": 0.000_000_05,
41
- "completion": 0.000_000_08,
42
- },
43
- "mixtral-8x7b-32768": {
44
- "prompt": 0.000_000_24,
45
- "completion": 0.000_000_24,
46
- },
47
- "gemma-7b-it": {
48
- "prompt": 0.000_000_07,
49
- "completion": 0.000_000_07,
50
- },
51
- "gemma2-9b-it": {
52
- "prompt": 0.000_000_2,
53
- "completion": 0.000_000_2,
54
- },
55
- }
56
-
57
- if input_tokens is None or output_tokens is None:
58
- return None
59
-
60
- try:
61
- model_pricing = pricing[model]
62
- except KeyError:
63
- return None
64
-
65
- prompt_cost = input_tokens * model_pricing["prompt"]
66
- completion_cost = output_tokens * model_pricing["completion"]
67
- total_cost = prompt_cost + completion_cost
68
-
69
- return total_cost
mirascope/core/mistral/_utils/_calculate_cost.py DELETED
@@ -1,48 +0,0 @@
1
- """Calculate the cost of a completion using the Mistral API."""
2
-
3
-
4
- def calculate_cost(
5
- input_tokens: int | float | None,
6
- cached_tokens: int | float | None,
7
- output_tokens: int | float | None,
8
- model: str = "open-mistral-7b",
9
- ) -> float | None:
10
- """Calculate the cost of a completion using the Mistral API.
11
-
12
- https://mistral.ai/technology/#pricing
13
-
14
- Model Input Cached Output
15
- open-mistral-nemo $0.3/1M tokens $0.3/1M tokens
16
- mistral-large-latest $3/1M tokens $9/1M tokens
17
- codestral-2405 $1/1M tokens $3/1M tokens
18
- open-mistral-7b $0.25/1M tokens $0.25/1M tokens
19
- open-mixtral-8x7b $0.7/1M tokens $0.7/1M tokens
20
- open-mixtral-8x22b $2/1M tokens $6/1M tokens
21
- mistral-small-latest $2/1M tokens $6/1M tokens
22
- mistral-medium-latest $2.75/1M tokens $8.1/1M tokens
23
- """
24
- pricing = {
25
- "open-mistral-nemo": {"prompt": 0.000_000_3, "completion": 0.000_000_3},
26
- "open-mistral-nemo-2407": {"prompt": 0.000_000_3, "completion": 0.000_000_3},
27
- "mistral-large-latest": {"prompt": 0.000_003, "completion": 0.000_009},
28
- "mistral-large-2407": {"prompt": 0.000_003, "completion": 0.000_009},
29
- "open-mistral-7b": {"prompt": 0.000_000_25, "completion": 0.000_000_25},
30
- "open-mixtral-8x7b": {"prompt": 0.000_000_7, "completion": 0.000_000_7},
31
- "open-mixtral-8x22b": {"prompt": 0.000_002, "completion": 0.000_006},
32
- "mistral-small-latest": {"prompt": 0.000_002, "completion": 0.000_006},
33
- "mistral-medium-latest": {"prompt": 0.000_002_75, "completion": 0.000_008_1},
34
- }
35
-
36
- if input_tokens is None or output_tokens is None:
37
- return None
38
-
39
- try:
40
- model_pricing = pricing[model]
41
- except KeyError:
42
- return None
43
-
44
- prompt_cost = input_tokens * model_pricing["prompt"]
45
- completion_cost = output_tokens * model_pricing["completion"]
46
- total_cost = prompt_cost + completion_cost
47
-
48
- return total_cost
mirascope/core/openai/_utils/_calculate_cost.py DELETED
@@ -1,246 +0,0 @@
1
- """Calculate the cost of a completion using the OpenAI API."""
2
-
3
-
4
- def calculate_cost(
5
- input_tokens: int | float | None,
6
- cached_tokens: int | float | None,
7
- output_tokens: int | float | None,
8
- model: str = "gpt-3.5-turbo-16k",
9
- ) -> float | None:
10
- """Calculate the cost of a completion using the OpenAI API.
11
-
12
- https://openai.com/pricing
13
-
14
- Model Input Cached Output
15
- gpt-4o $2.50 / 1M tokens $1.25 / 1M tokens $10.00 / 1M tokens
16
- gpt-4o-2024-11-20 $2.50 / 1M tokens $1.25 / 1M tokens $10.00 / 1M tokens
17
- gpt-4o-2024-08-06 $2.50 / 1M tokens $1.25 / 1M tokens $10.00 / 1M tokens
18
- gpt-4o-2024-05-13 $5.00 / 1M tokens $2.50 / 1M tokens $15.00 / 1M tokens
19
- gpt-4o-audio-preview $2.50 / 1M tokens $1.25 / 1M tokens $10.00 / 1M tokens
20
- gpt-4o-audio-preview-2024-12-17 $2.50 / 1M tokens $1.25 / 1M tokens $10.00 / 1M tokens
21
- gpt-4o-audio-preview-2024-10-01 $2.50 / 1M tokens $1.25 / 1M tokens $10.00 / 1M tokens
22
- gpt-4o-realtime-preview $5.00 / 1M tokens $2.50 / 1M tokens $20.00 / 1M tokens
23
- gpt-4o-realtime-preview-2024-12-17 $5.00 / 1M tokens $2.50 / 1M tokens $20.00 / 1M tokens
24
- gpt-4o-realtime-preview-2024-10-01 $5.00 / 1M tokens $2.50 / 1M tokens $20.00 / 1M tokens
25
- gpt-4o-mini $0.15 / 1M tokens $0.08 / 1M tokens $0.60 / 1M tokens
26
- gpt-4o-mini-2024-07-18 $0.15 / 1M tokens $0.08 / 1M tokens $0.60 / 1M tokens
27
- gpt-4o-mini-audio-preview $0.15 / 1M tokens $0.08 / 1M tokens $0.60 / 1M tokens
28
- gpt-4o-mini-audio-preview-2024-12-17 $0.15 / 1M tokens $0.08 / 1M tokens $0.60 / 1M tokens
29
- gpt-4o-mini-realtime-preview $0.60 / 1M tokens $0.30 / 1M tokens $2.40 / 1M tokens
30
- gpt-4o-mini-realtime-preview-2024-12-17 $0.60 / 1M tokens $0.30 / 1M tokens $2.40 / 1M tokens
31
- o1 $15.00 / 1M tokens $7.50 / 1M tokens $60.00 / 1M tokens
32
- o1-2024-12-17 $15.00 / 1M tokens $7.50 / 1M tokens $60.00 / 1M tokens
33
- o1-preview-2024-09-12 $15.00 / 1M tokens $7.50 / 1M tokens $60.00 / 1M tokens
34
- o3-mini $1.10 / 1M tokens $0.55 / 1M tokens $4.40 / 1M tokens
35
- o3-mini-2025-01-31 $1.10 / 1M tokens $0.55 / 1M tokens $4.40 / 1M tokens
36
- o1-mini $1.10 / 1M tokens $0.55 / 1M tokens $4.40 / 1M tokens
37
- o1-mini-2024-09-12 $1.10 / 1M tokens $0.55 / 1M tokens $4.40 / 1M tokens
38
- gpt-4-turbo $10.00 / 1M tokens $30.00 / 1M tokens
39
- gpt-4-turbo-2024-04-09 $10.00 / 1M tokens $30.00 / 1M tokens
40
- gpt-3.5-turbo-0125 $0.50 / 1M tokens $1.50 / 1M tokens
41
- gpt-3.5-turbo-1106 $1.00 / 1M tokens $2.00 / 1M tokens
42
- gpt-4-1106-preview $10.00 / 1M tokens $30.00 / 1M tokens
43
- gpt-4 $30.00 / 1M tokens $60.00 / 1M tokens
44
- text-embedding-3-small $0.02 / 1M tokens
45
- text-embedding-3-large $0.13 / 1M tokens
46
- text-embedding-ada-0002 $0.10 / 1M tokens
47
- """
48
- pricing = {
49
- "gpt-4o": {
50
- "prompt": 0.000_002_5,
51
- "cached": 0.000_001_25,
52
- "completion": 0.000_01,
53
- },
54
- "gpt-4o-2024-11-20": {
55
- "prompt": 0.000_002_5,
56
- "cached": 0.000_001_25,
57
- "completion": 0.000_01,
58
- },
59
- "gpt-4o-2024-08-06": {
60
- "prompt": 0.000_002_5,
61
- "cached": 0.000_001_25,
62
- "completion": 0.000_01,
63
- },
64
- "gpt-4o-2024-05-13": {
65
- "prompt": 0.000_005,
66
- "cached": 0.000_002_5,
67
- "completion": 0.000_015,
68
- },
69
- "gpt-4o-audio-preview": {
70
- "prompt": 0.000_002_5,
71
- "cached": 0.000_001_25,
72
- "completion": 0.000_01,
73
- },
74
- "gpt-4o-audio-preview-2024-12-17": {
75
- "prompt": 0.000_002_5,
76
- "cached": 0.000_001_25,
77
- "completion": 0.000_01,
78
- },
79
- "gpt-4o-audio-preview-2024-10-01": {
80
- "prompt": 0.000_002_5,
81
- "cached": 0.000_001_25,
82
- "completion": 0.000_01,
83
- },
84
- "gpt-4o-realtime-preview": {
85
- "prompt": 0.000_005,
86
- "cached": 0.000_002_5,
87
- "completion": 0.000_02,
88
- },
89
- "gpt-4o-realtime-preview-2024-12-17": {
90
- "prompt": 0.000_005,
91
- "cached": 0.000_002_5,
92
- "completion": 0.000_02,
93
- },
94
- "gpt-4o-realtime-preview-2024-10-01": {
95
- "prompt": 0.000_005,
96
- "cached": 0.000_002_5,
97
- "completion": 0.000_02,
98
- },
99
- "gpt-4o-mini": {
100
- "prompt": 0.000_000_15,
101
- "cached": 0.000_000_08,
102
- "completion": 0.000_000_6,
103
- },
104
- "gpt-4o-mini-2024-07-18": {
105
- "prompt": 0.000_000_15,
106
- "cached": 0.000_000_08,
107
- "completion": 0.000_000_6,
108
- },
109
- "gpt-4o-mini-audio-preview": {
110
- "prompt": 0.000_000_15,
111
- "cached": 0.000_000_08,
112
- "completion": 0.000_000_6,
113
- },
114
- "gpt-4o-mini-audio-preview-2024-12-17": {
115
- "prompt": 0.000_000_15,
116
- "cached": 0.000_000_08,
117
- "completion": 0.000_000_6,
118
- },
119
- "gpt-4o-mini-realtime-preview": {
120
- "prompt": 0.000_000_6,
121
- "cached": 0.000_000_3,
122
- "completion": 0.000_002_4,
123
- },
124
- "gpt-4o-mini-realtime-preview-2024-12-17": {
125
- "prompt": 0.000_000_6,
126
- "cached": 0.000_000_3,
127
- "completion": 0.000_002_4,
128
- },
129
- "o1": {
130
- "prompt": 0.000_015,
131
- "cached": 0.000_007_5,
132
- "completion": 0.000_06,
133
- },
134
- "o1-2024-12-17": {
135
- "prompt": 0.000_015,
136
- "cached": 0.000_007_5,
137
- "completion": 0.000_06,
138
- },
139
- "o1-preview-2024-09-12": {
140
- "prompt": 0.000_015,
141
- "cached": 0.000_007_5,
142
- "completion": 0.000_06,
143
- },
144
- "o3-mini": {
145
- "prompt": 0.000_001_1,
146
- "cached": 0.000_000_55,
147
- "completion": 0.000_004_4,
148
- },
149
- "o3-mini-2025-01-31": {
150
- "prompt": 0.000_001_1,
151
- "cached": 0.000_000_55,
152
- "completion": 0.000_004_4,
153
- },
154
- "o1-mini": {
155
- "prompt": 0.000_001_1,
156
- "cached": 0.000_000_55,
157
- "completion": 0.000_004_4,
158
- },
159
- "o1-mini-2024-09-12": {
160
- "prompt": 0.000_001_1,
161
- "cached": 0.000_000_55,
162
- "completion": 0.000_004_4,
163
- },
164
- "gpt-4-turbo": {
165
- "prompt": 0.000_01,
166
- "cached": 0,
167
- "completion": 0.000_03,
168
- },
169
- "gpt-4-turbo-2024-04-09": {
170
- "prompt": 0.000_01,
171
- "cached": 0,
172
- "completion": 0.000_03,
173
- },
174
- "gpt-3.5-turbo-0125": {
175
- "prompt": 0.000_000_5,
176
- "cached": 0,
177
- "completion": 0.000_001_5,
178
- },
179
- "gpt-3.5-turbo-1106": {
180
- "prompt": 0.000_001,
181
- "cached": 0,
182
- "completion": 0.000_002,
183
- },
184
- "gpt-4-1106-preview": {
185
- "prompt": 0.000_01,
186
- "cached": 0,
187
- "completion": 0.000_03,
188
- },
189
- "gpt-4": {
190
- "prompt": 0.000_003,
191
- "cached": 0,
192
- "completion": 0.000_006,
193
- },
194
- "gpt-3.5-turbo-4k": {
195
- "prompt": 0.000_015,
196
- "cached": 0,
197
- "completion": 0.000_02,
198
- },
199
- "gpt-3.5-turbo-16k": {
200
- "prompt": 0.000_003,
201
- "cached": 0,
202
- "completion": 0.000_004,
203
- },
204
- "gpt-4-8k": {
205
- "prompt": 0.000_003,
206
- "cached": 0,
207
- "completion": 0.000_006,
208
- },
209
- "gpt-4-32k": {
210
- "prompt": 0.000_006,
211
- "cached": 0,
212
- "completion": 0.000_012,
213
- },
214
- "text-embedding-3-small": {
215
- "prompt": 0.000_000_02,
216
- "cached": 0,
217
- "completion": 0,
218
- },
219
- "text-embedding-ada-002": {
220
- "prompt": 0.000_000_1,
221
- "cached": 0,
222
- "completion": 0,
223
- },
224
- "text-embedding-3-large": {
225
- "prompt": 0.000_000_13,
226
- "cached": 0,
227
- "completion": 0,
228
- },
229
- }
230
- if input_tokens is None or output_tokens is None:
231
- return None
232
-
233
- if cached_tokens is None:
234
- cached_tokens = 0
235
-
236
- try:
237
- model_pricing = pricing[model]
238
- except KeyError:
239
- return None
240
-
241
- prompt_cost = input_tokens * model_pricing["prompt"]
242
- cached_cost = cached_tokens * model_pricing["cached"]
243
- completion_cost = output_tokens * model_pricing["completion"]
244
- total_cost = prompt_cost + cached_cost + completion_cost
245
-
246
- return total_cost