Spaces:
Running
Running
File size: 5,264 Bytes
15a5288 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 |
// Core provider data structure based on provider-api-spec.md
export interface ProviderEntry {
provider: string;
status?: "live" | "offline" | "deprecated";
context_length?: number;
pricing?: {
input: number; // $ per 1M tokens
output: number; // $ per 1M tokens
};
quantization?: string;
max_completion_tokens?: number;
supported_parameters?: string[];
// Model identification
model_id?: string;
created?: number;
// Capability flags
supports_tools?: boolean;
supports_function_calling?: boolean;
supports_structured_output?: boolean;
supports_response_format?: boolean;
supports_streaming?: boolean;
supports_logprobs?: boolean;
supports_stop_sequences?: boolean;
supports_seed?: boolean;
supports_temperature?: boolean;
supports_top_p?: boolean;
supports_frequency_penalty?: boolean;
supports_presence_penalty?: boolean;
supports_repetition_penalty?: boolean;
supports_top_k?: boolean;
supports_min_p?: boolean;
supports_max_tokens?: boolean;
supports_logit_bias?: boolean;
supports_top_logprobs?: boolean;
supports_image_input?: boolean;
// Performance metrics
latency_s?: number;
throughput_tps?: number;
performance_error?: string;
performance_tested_at?: string;
// Additional metadata
owned_by?: string;
model_type?: string;
description?: string;
deprecated_at?: string;
model_class?: string;
is_gated?: boolean;
}
// Provider-specific response types
export interface GroqModel {
id: string;
object: string;
created: number;
owned_by: string;
active: boolean;
context_window: number;
public_apps: any;
max_completion_tokens: number;
}
export interface CohereModel {
name: string;
endpoints: string[];
finetuned: boolean;
context_length: number;
tokenizer_url: string;
supports_vision: boolean;
features: string[];
default_endpoints: string[];
is_deprecated?: boolean;
}
export interface FireworksModel {
id: string;
object: string;
owned_by: string;
created: number;
kind: string;
supports_chat: boolean;
supports_image_input: boolean;
supports_tools: boolean;
context_length: number;
}
export interface FireworksDetailedModel {
name: string;
displayName: string;
description: string;
contextLength: number;
baseModelDetails: {
checkpointFormat: string;
defaultPrecision: string;
modelType: string;
moe: boolean;
parameterCount: string;
supportsFireattention: boolean;
tunable: boolean;
worldSize: number;
};
defaultSamplingParams: {
temperature?: number;
top_p?: number;
max_tokens?: number;
[key: string]: any;
};
supportsImageInput: boolean;
supportsLora: boolean;
supportsTools: boolean;
state: string;
deprecationDate: string | null;
huggingFaceUrl: string;
supportedPrecisions: string[];
deployedModelRefs: any[];
}
export interface TogetherModel {
id: string;
object: string;
created: number;
type: string;
display_name: string;
organization: string;
context_length: number;
pricing: {
input: number; // $ per million tokens
output: number; // $ per million tokens
hourly: number;
base: number;
finetune: number;
};
config: {
chat_template: string;
stop: string[];
bos_token: string;
eos_token: string;
};
}
export interface SambaNovaModel {
id: string;
object: string;
owned_by: string;
context_length: number;
max_completion_tokens: number;
pricing: {
prompt: string; // $ per token
completion: string; // $ per token
};
sn_metadata: any;
}
export interface NovitaModel {
id: string;
object: string;
created: number;
owned_by: string;
input_token_price_per_m: number; // Cents per million tokens
output_token_price_per_m: number; // Cents per million tokens
title: string;
description: string;
context_size: number;
max_output_tokens: number;
model_type: string;
features: string[];
endpoints: string[];
status: number;
display_name: string;
}
export interface FeatherlessModel {
id: string;
is_gated: boolean;
created: number;
model_class: string;
owned_by: string;
context_length: number;
max_completion_tokens: number;
available_on_current_plan: boolean;
}
// Minimal response types
export interface CerebrasModel {
id: string;
object: string;
created: number;
owned_by: string;
}
export interface NebiusModel {
id: string;
created: number;
object: string;
owned_by: string;
}
export interface LambdaModel {
id: string;
object: string;
created: number;
owned_by: string;
}
// Base provider fetcher interface
export interface ProviderFetcher {
name: string;
fetchModels(): Promise<ProviderEntry[]>;
}
// Configuration for rate limiting
export interface RateLimitConfig {
requestsPerMinute?: number;
requestsPerHour?: number;
retryAttempts?: number;
initialBackoffMs?: number;
}
// Feature mapping types
export interface FeatureMapping {
[key: string]: keyof ProviderEntry | string[] | null;
}
// Static pricing data structure
export interface StaticPricing {
[provider: string]: {
[modelId: string]: {
input: number; // $ per 1M tokens
output: number; // $ per 1M tokens
};
};
} |