forgot files for DPAIS

This commit is contained in:
timothycarambat 2025-05-14 15:26:14 -07:00
parent 8560b0039f
commit 605910b76d
4 changed files with 513 additions and 0 deletions

View File

@ -0,0 +1,181 @@
import React, { useEffect, useState } from "react";
import { CaretDown, CaretUp } from "@phosphor-icons/react";
import System from "@/models/system";
import PreLoader from "@/components/Preloader";
import { DPAIS_COMMON_URLS } from "@/utils/constants";
import useProviderEndpointAutoDiscovery from "@/hooks/useProviderEndpointAutoDiscovery";
export default function DellProAIStudioOptions({
settings,
showAlert = false,
}) {
const {
autoDetecting: loading,
basePath,
basePathValue,
showAdvancedControls,
setShowAdvancedControls,
handleAutoDetectClick,
} = useProviderEndpointAutoDiscovery({
provider: "dpais",
initialBasePath: settings?.DellProAiStudioBasePath,
ENDPOINTS: DPAIS_COMMON_URLS,
});
return (
<div className="w-full flex flex-col gap-y-7">
<div className="w-full flex items-center gap-[36px] mt-1.5">
{!settings?.credentialsOnly && (
<>
<DellProAiStudioModelSelection
settings={settings}
basePath={basePath.value}
/>
<div className="flex flex-col w-60">
<label className="text-white text-sm font-semibold block mb-2">
Token context window
</label>
<input
type="number"
name="DellProAiStudioTokenLimit"
className="border-none bg-theme-settings-input-bg text-white placeholder:text-theme-settings-input-placeholder text-sm rounded-lg focus:outline-primary-button active:outline-primary-button outline-none block w-full p-2.5"
placeholder="4096"
min={1}
onScroll={(e) => e.target.blur()}
defaultValue={settings?.DellProAiStudioTokenLimit}
required={true}
autoComplete="off"
/>
</div>
</>
)}
</div>
<div className="flex justify-start mt-4">
<button
onClick={(e) => {
e.preventDefault();
setShowAdvancedControls(!showAdvancedControls);
}}
className="border-none text-theme-text-primary hover:text-theme-text-secondary flex items-center text-sm"
>
{showAdvancedControls ? "Hide" : "Show"} advanced settings
{showAdvancedControls ? (
<CaretUp size={14} className="ml-1" />
) : (
<CaretDown size={14} className="ml-1" />
)}
</button>
</div>
<div hidden={!showAdvancedControls}>
<div className="w-full flex items-center gap-4">
<div className="flex flex-col w-fit">
<div className="flex justify-between items-center mb-2 gap-x-2">
<label className="text-white text-sm font-semibold">
Dell Pro AI Studio Base URL
</label>
{loading ? (
<PreLoader size="6" />
) : (
<>
{!basePathValue.value && (
<button
onClick={handleAutoDetectClick}
className="bg-primary-button text-xs font-medium px-2 py-1 rounded-lg hover:bg-secondary hover:text-white shadow-[0_4px_14px_rgba(0,0,0,0.25)]"
>
Auto-Detect
</button>
)}
</>
)}
</div>
<input
type="url"
name="DellProAiStudioBasePath"
className="border-none bg-theme-settings-input-bg text-white placeholder:text-theme-settings-input-placeholder text-sm rounded-lg focus:outline-primary-button active:outline-primary-button outline-none block w-full p-2.5"
placeholder="http://localhost:8553/v1"
value={basePathValue.value}
required={true}
autoComplete="off"
spellCheck={false}
onChange={basePath.onChange}
onBlur={basePath.onBlur}
/>
</div>
</div>
</div>
</div>
);
}
function DellProAiStudioModelSelection({ settings, basePath = null }) {
const [customModels, setCustomModels] = useState([]);
const [loading, setLoading] = useState(true);
useEffect(() => {
async function findCustomModels() {
if (!basePath) {
setCustomModels([]);
setLoading(false);
return;
}
setLoading(true);
const { models } = await System.customModels(
"dpais",
null,
basePath,
2_000
);
setCustomModels(models || []);
setLoading(false);
}
findCustomModels();
}, [basePath]);
if (loading || customModels.length == 0) {
return (
<div className="flex flex-col w-60">
<label className="text-white text-sm font-semibold block mb-2">
Chat Model Selection
</label>
<select
name="DellProAiStudioModelPref"
disabled={true}
className="border-none bg-theme-settings-input-bg border-gray-500 text-white text-sm rounded-lg block w-full p-2.5"
>
<option disabled={true} selected={true}>
-- loading available models --
</option>
</select>
</div>
);
}
return (
<div className="flex flex-col w-60">
<label className="text-white text-sm font-semibold block mb-2">
Chat Model Selection
</label>
<select
name="DellProAiStudioModelPref"
required={true}
className="border-none bg-theme-settings-input-bg border-gray-500 text-white text-sm rounded-lg block w-full p-2.5"
>
{customModels.length > 0 && (
<optgroup label="Your loaded models">
{customModels.map((model) => {
return (
<option
key={model.id}
value={model.id}
selected={settings.DellProAiStudioModelPref === model.id}
>
{model.id}
</option>
);
})}
</optgroup>
)}
</select>
</div>
);
}

Binary file not shown.

After

Width:  |  Height:  |  Size: 9.9 KiB

View File

@ -0,0 +1,210 @@
const { NativeEmbedder } = require("../../EmbeddingEngines/native");
const {
handleDefaultStreamResponseV2,
formatChatHistory,
} = require("../../helpers/chat/responses");
const {
LLMPerformanceMonitor,
} = require("../../helpers/chat/LLMPerformanceMonitor");
// hybrid of openAi LLM chat completion for Dell Pro AI Studio
class DellProAiStudioLLM {
constructor(embedder = null, modelPreference = null) {
if (!process.env.DPAIS_LLM_BASE_PATH)
throw new Error("No Dell Pro AI Studio Base Path was set.");
const { OpenAI: OpenAIApi } = require("openai");
this.dpais = new OpenAIApi({
baseURL: DellProAiStudioLLM.parseBasePath(),
apiKey: null,
});
this.model = modelPreference || process.env.DPAIS_LLM_MODEL_PREF;
this.limits = {
history: this.promptWindowLimit() * 0.15,
system: this.promptWindowLimit() * 0.15,
user: this.promptWindowLimit() * 0.7,
};
this.embedder = embedder ?? new NativeEmbedder();
this.defaultTemp = 0.7;
this.log(
`Dell Pro AI Studio LLM initialized with ${this.model}. ctx: ${this.promptWindowLimit()}`
);
}
/**
* Parse the base path for the Dell Pro AI Studio API
* so we can use it for inference requests
* @param {string} providedBasePath
* @returns {string}
*/
static parseBasePath(providedBasePath = process.env.DPAIS_LLM_BASE_PATH) {
try {
const baseURL = new URL(providedBasePath);
const basePath = `${baseURL.origin}/v1/openai`;
return basePath;
} catch (e) {
return null;
}
}
log(text, ...args) {
console.log(`\x1b[36m[${this.constructor.name}]\x1b[0m ${text}`, ...args);
}
#appendContext(contextTexts = []) {
if (!contextTexts || !contextTexts.length) return "";
return (
"\nContext:\n" +
contextTexts
.map((text, i) => {
return `[CONTEXT ${i}]:\n${text}\n[END CONTEXT ${i}]\n\n`;
})
.join("")
);
}
streamingEnabled() {
return "streamGetChatCompletion" in this;
}
static promptWindowLimit(_modelName) {
const limit = process.env.DPAIS_LLM_MODEL_TOKEN_LIMIT || 4096;
if (!limit || isNaN(Number(limit)))
throw new Error("No Dell Pro AI Studio token context limit was set.");
return Number(limit);
}
// Ensure the user set a value for the token limit
// and if undefined - assume 4096 window.
promptWindowLimit() {
const limit = process.env.DPAIS_LLM_MODEL_TOKEN_LIMIT || 4096;
if (!limit || isNaN(Number(limit)))
throw new Error("No Dell Pro AI Studio token context limit was set.");
return Number(limit);
}
async isValidChatCompletionModel(_ = "") {
return true;
}
/**
* Generates appropriate content array for a message + attachments.
* @param {{userPrompt:string, attachments: import("../../helpers").Attachment[]}}
* @returns {string|object[]}
*/
#generateContent({ userPrompt, attachments = [] }) {
if (!attachments.length) return userPrompt;
const content = [{ type: "text", text: userPrompt }];
for (let attachment of attachments) {
content.push({
type: "image_url",
image_url: {
url: attachment.contentString,
detail: "auto",
},
});
}
return content.flat();
}
/**
* Construct the user prompt for this model.
* @param {{attachments: import("../../helpers").Attachment[]}} param0
* @returns
*/
constructPrompt({
systemPrompt = "",
contextTexts = [],
chatHistory = [],
userPrompt = "",
_attachments = [], // not used for Dell Pro AI Studio - `attachments` passed in is ignored
}) {
const prompt = {
role: "system",
content: `${systemPrompt}${this.#appendContext(contextTexts)}`,
};
return [
prompt,
...formatChatHistory(chatHistory, this.#generateContent),
{
role: "user",
content: this.#generateContent({ userPrompt, _attachments }),
},
];
}
async getChatCompletion(messages = null, { temperature = 0.7 }) {
if (!this.model)
throw new Error(
`Dell Pro AI Studio chat: ${this.model} is not valid or defined model for chat completion!`
);
const result = await LLMPerformanceMonitor.measureAsyncFunction(
this.dpais.chat.completions.create({
model: this.model,
messages,
temperature,
})
);
if (
!result.output.hasOwnProperty("choices") ||
result.output.choices.length === 0
)
return null;
return {
textResponse: result.output.choices[0].message.content,
metrics: {
prompt_tokens: result.output.usage?.prompt_tokens || 0,
completion_tokens: result.output.usage?.completion_tokens || 0,
total_tokens: result.output.usage?.total_tokens || 0,
outputTps: result.output.usage?.completion_tokens / result.duration,
duration: result.duration,
},
};
}
async streamGetChatCompletion(messages = null, { temperature = 0.7 }) {
if (!this.model)
throw new Error(
`Dell Pro AI Studio chat: ${this.model} is not valid or defined model for chat completion!`
);
const measuredStreamRequest = await LLMPerformanceMonitor.measureStream(
this.dpais.chat.completions.create({
model: this.model,
stream: true,
messages,
temperature,
}),
messages
);
return measuredStreamRequest;
}
handleStream(response, stream, responseProps) {
return handleDefaultStreamResponseV2(response, stream, responseProps);
}
// Simple wrapper for dynamic embedder & normalize interface for all LLM implementations
async embedTextInput(textInput) {
return await this.embedder.embedTextInput(textInput);
}
async embedChunks(textChunks = []) {
return await this.embedder.embedChunks(textChunks);
}
async compressMessages(promptArgs = {}, rawHistory = []) {
const { messageArrayCompressor } = require("../../helpers/chat");
const messageArray = this.constructPrompt(promptArgs);
return await messageArrayCompressor(this, messageArray, rawHistory);
}
}
module.exports = {
DellProAiStudioLLM,
};

View File

@ -0,0 +1,122 @@
const OpenAI = require("openai");
const Provider = require("./ai-provider.js");
const InheritMultiple = require("./helpers/classes.js");
const UnTooled = require("./helpers/untooled.js");
const {
DellProAiStudioLLM,
} = require("../../../AiProviders/dellProAiStudio/index.js");
/**
* The agent provider for Dell Pro AI Studio.
*/
class DellProAiStudioProvider extends InheritMultiple([Provider, UnTooled]) {
model;
/**
*
* @param {{model?: string}} config
*/
constructor(config = {}) {
super();
const model = config?.model || process.env.DPAIS_LLM_MODEL_PREF;
const client = new OpenAI({
baseURL: DellProAiStudioLLM.parseBasePath(), // Will use process.env.DPAIS_LLM_BASE_PATH if not provided
apiKey: null,
});
this._client = client;
this.model = model;
this.verbose = true;
}
get client() {
return this._client;
}
async #handleFunctionCallChat({ messages = [] }) {
return await this.client.chat.completions
.create({
model: this.model,
messages,
})
.then((result) => {
if (!result.hasOwnProperty("choices"))
throw new Error("DellProAiStudio chat: No results!");
if (result.choices.length === 0)
throw new Error("DellProAiStudio chat: No results length!");
return result.choices[0].message.content;
})
.catch((_) => {
return null;
});
}
/**
* Create a completion based on the received messages.
*
* @param messages A list of messages to send to the API.
* @param functions
* @returns The completion.
*/
async complete(messages, functions = []) {
try {
let completion;
if (functions.length > 0) {
const { toolCall, text } = await this.functionCall(
messages,
functions,
this.#handleFunctionCallChat.bind(this)
);
if (toolCall !== null) {
this.providerLog(`Valid tool call found - running ${toolCall.name}.`);
this.deduplicator.trackRun(toolCall.name, toolCall.arguments);
return {
result: null,
functionCall: {
name: toolCall.name,
arguments: toolCall.arguments,
},
cost: 0,
};
}
completion = { content: text };
}
if (!completion?.content) {
this.providerLog(
"Will assume chat completion without tool call inputs."
);
const response = await this.client.chat.completions.create({
model: this.model,
messages: this.cleanMsgs(messages),
});
completion = response.choices[0].message;
}
// The UnTooled class inherited Deduplicator is mostly useful to prevent the agent
// from calling the exact same function over and over in a loop within a single chat exchange
// _but_ we should enable it to call previously used tools in a new chat interaction.
this.deduplicator.reset("runs");
return {
result: completion.content,
cost: 0,
};
} catch (error) {
throw error;
}
}
/**
* Get the cost of the completion.
*
* @param _usage The completion to get the cost for.
* @returns The cost of the completion.
* Stubbed since LMStudio has no cost basis.
*/
getCost(_usage) {
return 0;
}
}
module.exports = DellProAiStudioProvider;