File size: 3,343 Bytes
b924465 |
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 |
import type { GenerationConfig } from "$lib/components/InferencePlayground/generationConfigSettings";
import type { ChatCompletionInputMessage } from "@huggingface/tasks";
export type ConversationMessage = Omit<ChatCompletionInputMessage, "content"> & { content?: string };
export type Conversation = {
model: ModelWithTokenizer;
config: GenerationConfig;
messages: ConversationMessage[];
systemMessage: ConversationMessage;
streaming: boolean;
provider?: string;
};
export type Session = {
conversations: [Conversation] | [Conversation, Conversation];
};
interface TokenizerConfig {
chat_template?: string;
model_max_length?: number;
}
export type ModelWithTokenizer = Model & {
tokenizerConfig: TokenizerConfig;
};
export type Model = {
_id: string;
id: string;
inferenceProviderMapping: InferenceProviderMapping[];
trendingScore: number;
config: Config;
tags: string[];
pipeline_tag: PipelineTag;
library_name?: LibraryName;
};
export type Config = {
architectures: string[];
model_type: string;
tokenizer_config: TokenizerConfig;
auto_map?: AutoMap;
quantization_config?: QuantizationConfig;
};
export type AutoMap = {
AutoConfig: string;
AutoModel?: string;
AutoModelForCausalLM: string;
AutoModelForSequenceClassification?: string;
AutoModelForTokenClassification?: string;
AutoModelForQuestionAnswering?: string;
};
export type QuantizationConfig = {
quant_method: string;
bits?: number;
};
// export type TokenizerConfig = {
// bos_token?: Token | BosTokenEnum | null;
// chat_template: ChatTemplateElement[] | string;
// eos_token: Token | EOSTokenEnum;
// pad_token?: Token | null | string;
// unk_token?: Token | UnkTokenEnum | null;
// use_default_system_prompt?: boolean;
// };
export type Token = {
__type: Type;
content: Content;
lstrip: boolean;
normalized: boolean;
rstrip: boolean;
single_word: boolean;
};
export enum Type {
AddedToken = "AddedToken",
}
export enum Content {
BeginOfSentence = "<｜begin▁of▁sentence｜>",
ContentS = "</s>",
EndOfSentence = "<｜end▁of▁sentence｜>",
S = "<s>",
Unk = "<unk>",
}
export enum BosTokenEnum {
BeginOfText = "<|begin_of_text|>",
Bos = "<bos>",
BosToken = "<BOS_TOKEN>",
Endoftext = "<|endoftext|>",
IMStart = "<|im_start|>",
S = "<s>",
Startoftext = "<|startoftext|>",
}
export type ChatTemplateElement = {
name: string;
template: string;
};
export enum EOSTokenEnum {
EOS = "<eos>",
EndOfText = "<|end_of_text|>",
EndOfTurnToken = "<|END_OF_TURN_TOKEN|>",
Endoftext = "<|endoftext|>",
EotID = "<|eot_id|>",
IMEnd = "<|im_end|>",
S = "</s>",
}
export enum UnkTokenEnum {
Endoftext = "<|endoftext|>",
Unk = "<unk>",
}
export type InferenceProviderMapping = {
provider: Provider;
providerId: string;
status: Status;
task: Task;
};
export enum Provider {
Cerebras = "cerebras",
FalAI = "fal-ai",
FireworksAI = "fireworks-ai",
HFInference = "hf-inference",
Hyperbolic = "hyperbolic",
Nebius = "nebius",
Novita = "novita",
Replicate = "replicate",
Sambanova = "sambanova",
Together = "together",
}
export enum Status {
Live = "live",
Staging = "staging",
}
export enum Task {
Conversational = "conversational",
}
export enum LibraryName {
Mlx = "mlx",
Transformers = "transformers",
Vllm = "vllm",
}
export enum PipelineTag {
TextGeneration = "text-generation",
}
|