Performance enhancements galore! New themining system

This is once again a huge commit, but its mostly performance
improvements along with some bug fixes and refactoring. It also includes
changes to the theming systems. I'm still not 100% happy with the
theming system, but its better than before.

Model fetching has been dramatically improved! Nearly all the important
computation and pre-processing has been moved to the server. This has
also somehow fixed the way model details are loaded, which was causing
many models to be missing their details despite models.dev having them.

The markdown renderer has once again been changed, but I'm mostly
certain that this is the last time major changes will be made to it. The
renderer is not spamming components, bloating memory usage, and its not
using a bug prone custom written chunking system.

There's also a lot more that I haven't mentioned and honestly forgot. I
need to get better commit hygiene tbh.
This commit is contained in:
Zoe
2026-02-19 23:44:23 -06:00
parent 32a4f7f95d
commit 59bb7fbc12
85 changed files with 3523 additions and 2039 deletions
+33 -9
View File
@@ -9,7 +9,8 @@ import { httpClient } from '~~/server/lib/triplit';
import { addPendingGeneration, completeGeneration } from '~~/server/utils/generations';
import type { schema } from '~~/triplit/schema';
import { spawn } from 'child_process';
import { getGateway, type ModelGateway } from '~~/server/utils/ai-provider';
import { GatewayFetchError, getProviderDetails, type ModelGateway } from '~~/server/utils/ai-provider';
import { assert } from '~~/utils/assert';
export const messagesSchema = z.array(modelMessageSchema);
@@ -73,7 +74,26 @@ export default defineEventHandler(async (event) => {
});
}
const { gateway, streamTransformer: transformer } = await getGateway(provider, model, providerApiKey);
const providerDetails = await getProviderDetails(provider, providerApiKey, model);
if (!providerDetails.ok) {
switch (providerDetails.error) {
case GatewayFetchError.NoProviderApiKey: {
throw createError({
statusCode: 400,
message: `${provider.type} provider requires an API key`,
});
}
case GatewayFetchError.NoProviderBaseUrl: {
throw createError({
statusCode: 400,
message: 'Invalid provider URL',
});
}
}
}
const { gateway } = providerDetails.data;
assert(gateway !== null, 'Invalid gateway');
const generationId = nanoid();
const message = await httpClient.insert('messages', {
@@ -98,14 +118,20 @@ export default defineEventHandler(async (event) => {
let logMessage: ((message: string) => void) | undefined;
if (process.env.GENERATION_DEBUG) {
logFile = await fs.open(path.join(process.env.LOG_DIR!, `${Date.now()}-${generationId}.log`), 'w');
logMessage = (message: string) => {
logFile!.write(message + '\n');
};
if (process.env.LOG_DIR) {
await fs.mkdir(process.env.LOG_DIR!, { recursive: true });
logFile = await fs.open(path.join(process.env.LOG_DIR!, `${Date.now()}-${generationId}.log`), 'w');
logMessage = (message: string) => {
logFile!.write(message + '\n');
};
} else {
console.warn('Generation debug logging is enabled but GENERATION_DEBUG is not set');
}
}
event.waitUntil(
generateResponse(message, { gateway, model, parameters: args }, generationId, userId, topicId, messages, transformer, logMessage, logFile),
generateResponse(message, { gateway: gateway.gateway, model, parameters: args }, generationId, userId, topicId, messages, gateway.streamTransformer, logMessage, logFile),
);
return {
@@ -755,8 +781,6 @@ async function generateResponse(
status: 'failed',
error,
});
throw new Error(error);
} break;
case 'finish': {
let tps;