File size: 7,561 Bytes
0ae355f | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 | from datetime import date
from enum import Enum
from typing import Any
from pydantic import BaseModel, Field
class GeminiSafetyCategory(str, Enum):
HARM_CATEGORY_SEXUALLY_EXPLICIT = "HARM_CATEGORY_SEXUALLY_EXPLICIT"
HARM_CATEGORY_HATE_SPEECH = "HARM_CATEGORY_HATE_SPEECH"
HARM_CATEGORY_HARASSMENT = "HARM_CATEGORY_HARASSMENT"
HARM_CATEGORY_DANGEROUS_CONTENT = "HARM_CATEGORY_DANGEROUS_CONTENT"
class GeminiSafetyThreshold(str, Enum):
OFF = "OFF"
BLOCK_NONE = "BLOCK_NONE"
BLOCK_LOW_AND_ABOVE = "BLOCK_LOW_AND_ABOVE"
BLOCK_MEDIUM_AND_ABOVE = "BLOCK_MEDIUM_AND_ABOVE"
BLOCK_ONLY_HIGH = "BLOCK_ONLY_HIGH"
class GeminiSafetySetting(BaseModel):
category: GeminiSafetyCategory
threshold: GeminiSafetyThreshold
class GeminiRole(str, Enum):
user = "user"
model = "model"
class GeminiMimeType(str, Enum):
application_pdf = "application/pdf"
audio_mpeg = "audio/mpeg"
audio_mp3 = "audio/mp3"
audio_wav = "audio/wav"
image_png = "image/png"
image_jpeg = "image/jpeg"
image_webp = "image/webp"
text_plain = "text/plain"
video_mov = "video/mov"
video_mpeg = "video/mpeg"
video_mp4 = "video/mp4"
video_mpg = "video/mpg"
video_avi = "video/avi"
video_wmv = "video/wmv"
video_mpegps = "video/mpegps"
video_flv = "video/flv"
class GeminiInlineData(BaseModel):
data: str | None = Field(
None,
description="The base64 encoding of the image, PDF, or video to include inline in the prompt. "
"When including media inline, you must also specify the media type (mimeType) of the data. Size limit: 20MB",
)
mimeType: GeminiMimeType | None = Field(None)
class GeminiFileData(BaseModel):
fileUri: str | None = Field(None)
mimeType: GeminiMimeType | None = Field(None)
class GeminiPart(BaseModel):
inlineData: GeminiInlineData | None = Field(None)
fileData: GeminiFileData | None = Field(None)
text: str | None = Field(None)
class GeminiTextPart(BaseModel):
text: str | None = Field(None)
class GeminiContent(BaseModel):
parts: list[GeminiPart] = Field([])
role: GeminiRole = Field(..., examples=["user"])
class GeminiSystemInstructionContent(BaseModel):
parts: list[GeminiTextPart] = Field(
...,
description="A list of ordered parts that make up a single message. "
"Different parts may have different IANA MIME types.",
)
role: GeminiRole | None = Field(..., description="The role field of systemInstruction may be ignored.")
class GeminiFunctionDeclaration(BaseModel):
description: str | None = Field(None)
name: str = Field(...)
parameters: dict[str, Any] = Field(..., description="JSON schema for the function parameters")
class GeminiTool(BaseModel):
functionDeclarations: list[GeminiFunctionDeclaration] | None = Field(None)
class GeminiOffset(BaseModel):
nanos: int | None = Field(None, ge=0, le=999999999)
seconds: int | None = Field(None, ge=-315576000000, le=315576000000)
class GeminiVideoMetadata(BaseModel):
endOffset: GeminiOffset | None = Field(None)
startOffset: GeminiOffset | None = Field(None)
class GeminiGenerationConfig(BaseModel):
maxOutputTokens: int | None = Field(None, ge=16, le=8192)
seed: int | None = Field(None)
stopSequences: list[str] | None = Field(None)
temperature: float | None = Field(None, ge=0.0, le=2.0)
topK: int | None = Field(None, ge=1)
topP: float | None = Field(None, ge=0.0, le=1.0)
class GeminiImageConfig(BaseModel):
aspectRatio: str | None = Field(None)
imageSize: str | None = Field(None)
class GeminiImageGenerationConfig(GeminiGenerationConfig):
responseModalities: list[str] | None = Field(None)
imageConfig: GeminiImageConfig | None = Field(None)
class GeminiImageGenerateContentRequest(BaseModel):
contents: list[GeminiContent] = Field(...)
generationConfig: GeminiImageGenerationConfig | None = Field(None)
safetySettings: list[GeminiSafetySetting] | None = Field(None)
systemInstruction: GeminiSystemInstructionContent | None = Field(None)
tools: list[GeminiTool] | None = Field(None)
videoMetadata: GeminiVideoMetadata | None = Field(None)
uploadImagesToStorage: bool = Field(True)
class GeminiGenerateContentRequest(BaseModel):
contents: list[GeminiContent] = Field(...)
generationConfig: GeminiGenerationConfig | None = Field(None)
safetySettings: list[GeminiSafetySetting] | None = Field(None)
systemInstruction: GeminiSystemInstructionContent | None = Field(None)
tools: list[GeminiTool] | None = Field(None)
videoMetadata: GeminiVideoMetadata | None = Field(None)
class Modality(str, Enum):
MODALITY_UNSPECIFIED = "MODALITY_UNSPECIFIED"
TEXT = "TEXT"
IMAGE = "IMAGE"
VIDEO = "VIDEO"
AUDIO = "AUDIO"
DOCUMENT = "DOCUMENT"
class ModalityTokenCount(BaseModel):
modality: Modality | None = None
tokenCount: int | None = Field(None, description="Number of tokens for the given modality.")
class Probability(str, Enum):
NEGLIGIBLE = "NEGLIGIBLE"
LOW = "LOW"
MEDIUM = "MEDIUM"
HIGH = "HIGH"
UNKNOWN = "UNKNOWN"
class GeminiSafetyRating(BaseModel):
category: GeminiSafetyCategory | None = None
probability: Probability | None = Field(
None,
description="The probability that the content violates the specified safety category",
)
class GeminiCitation(BaseModel):
authors: list[str] | None = None
endIndex: int | None = None
license: str | None = None
publicationDate: date | None = None
startIndex: int | None = None
title: str | None = None
uri: str | None = None
class GeminiCitationMetadata(BaseModel):
citations: list[GeminiCitation] | None = None
class GeminiCandidate(BaseModel):
citationMetadata: GeminiCitationMetadata | None = None
content: GeminiContent | None = None
finishReason: str | None = None
safetyRatings: list[GeminiSafetyRating] | None = None
class GeminiPromptFeedback(BaseModel):
blockReason: str | None = None
blockReasonMessage: str | None = None
safetyRatings: list[GeminiSafetyRating] | None = None
class GeminiUsageMetadata(BaseModel):
cachedContentTokenCount: int | None = Field(
None,
description="Output only. Number of tokens in the cached part in the input (the cached content).",
)
candidatesTokenCount: int | None = Field(None, description="Number of tokens in the response(s).")
candidatesTokensDetails: list[ModalityTokenCount] | None = Field(
None, description="Breakdown of candidate tokens by modality."
)
promptTokenCount: int | None = Field(
None,
description="Number of tokens in the request. When cachedContent is set, this is still the total effective prompt size meaning this includes the number of tokens in the cached content.",
)
promptTokensDetails: list[ModalityTokenCount] | None = Field(
None, description="Breakdown of prompt tokens by modality."
)
thoughtsTokenCount: int | None = Field(None, description="Number of tokens present in thoughts output.")
toolUsePromptTokenCount: int | None = Field(None, description="Number of tokens present in tool-use prompt(s).")
class GeminiGenerateContentResponse(BaseModel):
candidates: list[GeminiCandidate] | None = Field(None)
promptFeedback: GeminiPromptFeedback | None = Field(None)
usageMetadata: GeminiUsageMetadata | None = Field(None)
modelVersion: str | None = Field(None)
|