support glm-4v-flash

app/client/platforms/openai.ts

@@ -240,7 +240,7 @@ export class ChatGPTApi implements LLMApi {
       }
 
       // add max_tokens to vision model
-      if (visionModel) {
+      if (visionModel && modelConfig.model !== "glm-4v-flash") {
         requestPayload["max_tokens"] = Math.max(modelConfig.max_tokens, 4000);
       }
     }

app/utils.ts

@@ -265,6 +265,7 @@ export function isVisionModel(model: string) {
     "learnlm",
     "qwen-vl",
     "qwen2-vl",
+    "glm-4v",
   ];
   const isGpt4Turbo =
     model.includes("gpt-4-turbo") && !model.includes("preview");