Mirrors
/
oai-reverse-proxy
mirror of https://gitgud.io/khanon/oai-reverse-proxy.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

adds chatgpt-4o-latest

This commit is contained in:
nai-degen 2024-08-15 11:54:42 -05:00
parent f7fd5f00f2
commit 5e1fffe07d
5 changed files with 10 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
src/config.ts
View File

@ -415,18 +415,18 @@ export const config: Config = {
firebaseKey: getEnvWithDefault("FIREBASE_KEY", undefined), firebaseKey: getEnvWithDefault("FIREBASE_KEY", undefined),
textModelRateLimit: getEnvWithDefault("TEXT_MODEL_RATE_LIMIT", 4), textModelRateLimit: getEnvWithDefault("TEXT_MODEL_RATE_LIMIT", 4),
imageModelRateLimit: getEnvWithDefault("IMAGE_MODEL_RATE_LIMIT", 4), imageModelRateLimit: getEnvWithDefault("IMAGE_MODEL_RATE_LIMIT", 4),
maxContextTokensOpenAI: getEnvWithDefault("MAX_CONTEXT_TOKENS_OPENAI", 16384), maxContextTokensOpenAI: getEnvWithDefault("MAX_CONTEXT_TOKENS_OPENAI", 32768),
maxContextTokensAnthropic: getEnvWithDefault( maxContextTokensAnthropic: getEnvWithDefault(
"MAX_CONTEXT_TOKENS_ANTHROPIC", "MAX_CONTEXT_TOKENS_ANTHROPIC",
0 32768
), ),
maxOutputTokensOpenAI: getEnvWithDefault( maxOutputTokensOpenAI: getEnvWithDefault(
["MAX_OUTPUT_TOKENS_OPENAI", "MAX_OUTPUT_TOKENS"], ["MAX_OUTPUT_TOKENS_OPENAI", "MAX_OUTPUT_TOKENS"],
400 1024
), ),
maxOutputTokensAnthropic: getEnvWithDefault( maxOutputTokensAnthropic: getEnvWithDefault(
["MAX_OUTPUT_TOKENS_ANTHROPIC", "MAX_OUTPUT_TOKENS"], ["MAX_OUTPUT_TOKENS_ANTHROPIC", "MAX_OUTPUT_TOKENS"],
400 1024
), ),
allowedModelFamilies: getEnvWithDefault( allowedModelFamilies: getEnvWithDefault(
"ALLOWED_MODEL_FAMILIES", "ALLOWED_MODEL_FAMILIES",

2
src/proxy/middleware/request/preprocessors/validate-context-size.ts
View File

@ -58,6 +58,8 @@ export const validateContextSize: RequestPreprocessor = async (req) => {
modelMax = 16384; modelMax = 16384;
} else if (model.match(/^gpt-4o/)) { } else if (model.match(/^gpt-4o/)) {
modelMax = 128000; modelMax = 128000;
} else if (model.match(/^chatgpt-4o/)) {
modelMax = 128000;
} else if (model.match(/gpt-4-turbo(-\d{4}-\d{2}-\d{2})?$/)) { } else if (model.match(/gpt-4-turbo(-\d{4}-\d{2}-\d{2})?$/)) {
modelMax = 131072; modelMax = 131072;
} else if (model.match(/gpt-4-turbo(-preview)?$/)) { } else if (model.match(/gpt-4-turbo(-preview)?$/)) {

2
src/proxy/openai.ts
View File

@ -35,6 +35,8 @@ export const KNOWN_OPENAI_MODELS = [
// GPT4o Mini // GPT4o Mini
"gpt-4o-mini", "gpt-4o-mini",
"gpt-4o-mini-2024-07-18", "gpt-4o-mini-2024-07-18",
// GPT4o (ChatGPT)
"chatgpt-4o-latest",
// GPT4 Turbo (superceded by GPT4o) // GPT4 Turbo (superceded by GPT4o)
"gpt-4-turbo", "gpt-4-turbo",
"gpt-4-turbo-2024-04-09", // gpt4-turbo stable, with vision "gpt-4-turbo-2024-04-09", // gpt4-turbo stable, with vision

2
src/shared/api-schemas/openai.ts
View File

@ -52,7 +52,7 @@ export const OpenAIV1ChatCompletionSchema = z
.number() .number()
.int() .int()
.nullish() .nullish()
.default(Math.min(OPENAI_OUTPUT_MAX, 4096)) .default(Math.min(OPENAI_OUTPUT_MAX, 16384))
.transform((v) => Math.min(v ?? OPENAI_OUTPUT_MAX, OPENAI_OUTPUT_MAX)), .transform((v) => Math.min(v ?? OPENAI_OUTPUT_MAX, OPENAI_OUTPUT_MAX)),
frequency_penalty: z.number().optional().default(0), frequency_penalty: z.number().optional().default(0),
presence_penalty: z.number().optional().default(0), presence_penalty: z.number().optional().default(0),

1
src/shared/models.ts
View File

@ -130,6 +130,7 @@ export const IMAGE_GEN_MODELS: ModelFamily[] = ["dall-e", "azure-dall-e"];
export const OPENAI_MODEL_FAMILY_MAP: { [regex: string]: OpenAIModelFamily } = { export const OPENAI_MODEL_FAMILY_MAP: { [regex: string]: OpenAIModelFamily } = {
"^gpt-4o(-\\d{4}-\\d{2}-\\d{2})?$": "gpt4o", "^gpt-4o(-\\d{4}-\\d{2}-\\d{2})?$": "gpt4o",
"^chatgpt-4o": "gpt4o",
"^gpt-4o-mini(-\\d{4}-\\d{2}-\\d{2})?$": "turbo", // closest match "^gpt-4o-mini(-\\d{4}-\\d{2}-\\d{2})?$": "turbo", // closest match
"^gpt-4-turbo(-\\d{4}-\\d{2}-\\d{2})?$": "gpt4-turbo", "^gpt-4-turbo(-\\d{4}-\\d{2}-\\d{2})?$": "gpt4-turbo",
"^gpt-4-turbo(-preview)?$": "gpt4-turbo", "^gpt-4-turbo(-preview)?$": "gpt4-turbo",