chat: remember last model

server/src/llm/multiplexer.ts

@@ -10,11 +10,12 @@ function asProviderEnum(p: Provider) {
 
 export async function runMultiplex(req: MultiplexRequest): Promise<MultiplexResponse> {
   const t0 = performance.now();
+  const chatId = req.chatId ?? (await prisma.chat.create({ data: {}, select: { id: true } })).id;
 
   // Persist call record early so we can attach errors.
   const call = await prisma.llmCall.create({
     data: {
-      chatId: req.chatId ?? (await prisma.chat.create({ data: {} })).id,
+      chatId,
       provider: asProviderEnum(req.provider) as any,
       model: req.model,
       request: req as any,
@@ -22,6 +23,23 @@ export async function runMultiplex(req: MultiplexRequest): Promise<MultiplexResp
     select: { id: true, chatId: true },
   });
 
+  await prisma.$transaction([
+    prisma.chat.update({
+      where: { id: chatId },
+      data: {
+        lastUsedProvider: asProviderEnum(req.provider) as any,
+        lastUsedModel: req.model,
+      },
+    }),
+    prisma.chat.updateMany({
+      where: { id: chatId, initiatedProvider: null },
+      data: {
+        initiatedProvider: asProviderEnum(req.provider) as any,
+        initiatedModel: req.model,
+      },
+    }),
+  ]);
+
   try {
     let outText = "";
     let usage: MultiplexResponse["usage"] | undefined;

server/src/llm/streaming.ts

@@ -29,6 +29,23 @@ export async function* runMultiplexStream(req: MultiplexRequest): AsyncGenerator
     select: { id: true },
   });
 
+  await prisma.$transaction([
+    prisma.chat.update({
+      where: { id: chatId },
+      data: {
+        lastUsedProvider: req.provider as any,
+        lastUsedModel: req.model,
+      },
+    }),
+    prisma.chat.updateMany({
+      where: { id: chatId, initiatedProvider: null },
+      data: {
+        initiatedProvider: req.provider as any,
+        initiatedModel: req.model,
+      },
+    }),
+  ]);
+
   yield { type: "meta", chatId, callId: call.id, provider: req.provider, model: req.model };
 
   let text = "";