add provider field to chats (#4959)

2026-02-02 20:11:18 -08:00 · 2026-02-02 20:11:18 -08:00 · fef5bf06ec
commit fef5bf06ec
parent 8e2efc63d0
36 changed files with 83 additions and 23 deletions
--- a/server/utils/AiProviders/anthropic/index.js
+++ b/server/utils/AiProviders/anthropic/index.js
@ -172,6 +172,7 @@ class AnthropicLLM {
          outputTps: completionTokens / result.duration,
          duration: result.duration,
          model: this.model,
+          provider: this.className,
          timestamp: new Date(),
        },
      };
@ -194,6 +195,7 @@ class AnthropicLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/apipie/index.js
+++ b/server/utils/AiProviders/apipie/index.js
@ -221,6 +221,7 @@ class ApiPieLLM {
          (result.output.usage?.completion_tokens || 0) / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -242,6 +243,7 @@ class ApiPieLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/azureOpenAi/index.js
+++ b/server/utils/AiProviders/azureOpenAi/index.js
@ -15,6 +15,7 @@ class AzureOpenAiLLM {
    if (!process.env.AZURE_OPENAI_KEY)
      throw new Error("No Azure API key was set.");

+    this.className = "AzureOpenAiLLM";
    this.openai = new OpenAI({
      apiKey: process.env.AZURE_OPENAI_KEY,
      baseURL: AzureOpenAiLLM.formatBaseUrl(process.env.AZURE_OPENAI_ENDPOINT),
@ -175,6 +176,7 @@ class AzureOpenAiLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -197,6 +199,7 @@ class AzureOpenAiLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/bedrock/index.js
+++ b/server/utils/AiProviders/bedrock/index.js
@ -72,6 +72,7 @@ class AWSBedrockLLM {
        throw new Error(`Required environment variable ${envVar} is not set.`);
    }

+    this.className = "AWSBedrockLLM";
    this.model =
      modelPreference || process.env.AWS_BEDROCK_LLM_MODEL_PREFERENCE;

@ -448,6 +449,7 @@ class AWSBedrockLLM {
        outputTps: outputTps,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -493,6 +495,7 @@ class AWSBedrockLLM {
        messages,
        runPromptTokenCalculation: false,
        modelTag: this.model,
+        provider: this.className,
      });
      return measuredStreamRequest;
    } catch (e) {
--- a/server/utils/AiProviders/cohere/index.js
+++ b/server/utils/AiProviders/cohere/index.js
@ -125,6 +125,7 @@ class CohereLLM {
        outputTps: completionTokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -143,6 +144,7 @@ class CohereLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/cometapi/index.js
+++ b/server/utils/AiProviders/cometapi/index.js
@ -226,6 +226,7 @@ class CometApiLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -247,6 +248,7 @@ class CometApiLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/deepseek/index.js
+++ b/server/utils/AiProviders/deepseek/index.js
@ -131,6 +131,7 @@ class DeepSeekLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -152,6 +153,7 @@ class DeepSeekLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/dellProAiStudio/index.js
+++ b/server/utils/AiProviders/dellProAiStudio/index.js
@ -166,6 +166,7 @@ class DellProAiStudioLLM {
        outputTps: result.output.usage?.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -187,6 +188,7 @@ class DellProAiStudioLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/dockerModelRunner/index.js
+++ b/server/utils/AiProviders/dockerModelRunner/index.js
@ -26,6 +26,7 @@ class DockerModelRunnerLLM {
    if (!process.env.DOCKER_MODEL_RUNNER_LLM_MODEL_PREF && !modelPreference)
      throw new Error("No Docker Model Runner Model Pref was set.");

+    this.className = "DockerModelRunnerLLM";
    this.dmr = new OpenAIApi({
      baseURL: parseDockerModelRunnerEndpoint(
        process.env.DOCKER_MODEL_RUNNER_BASE_PATH
@ -173,6 +174,7 @@ class DockerModelRunnerLLM {
        outputTps: result.output.usage?.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -194,6 +196,7 @@ class DockerModelRunnerLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/fireworksAi/index.js
+++ b/server/utils/AiProviders/fireworksAi/index.js
@ -164,6 +164,7 @@ class FireworksAiLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -185,6 +186,7 @@ class FireworksAiLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/foundry/index.js
+++ b/server/utils/AiProviders/foundry/index.js
@ -234,6 +234,7 @@ class FoundryLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -256,6 +257,7 @@ class FoundryLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/gemini/index.js
+++ b/server/utils/AiProviders/gemini/index.js
@ -406,6 +406,7 @@ class GeminiLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -425,6 +426,7 @@ class GeminiLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/genericOpenAi/index.js
+++ b/server/utils/AiProviders/genericOpenAi/index.js
@ -194,6 +194,7 @@ class GenericOpenAiLLM {
          (result.output?.usage?.completion_tokens || 0) / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -209,9 +210,9 @@ class GenericOpenAiLLM {
        max_tokens: this.maxTokens,
      }),
      messages,
-      // runPromptTokenCalculation: true - There is not way to know if the generic provider connected is returning
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/giteeai/index.js
+++ b/server/utils/AiProviders/giteeai/index.js
@ -23,6 +23,7 @@ class GiteeAILLM {
      throw new Error("No Gitee AI API key was set.");
    const { OpenAI: OpenAIApi } = require("openai");

+    this.className = "GiteeAILLM";
    this.openai = new OpenAIApi({
      apiKey: process.env.GITEE_AI_API_KEY,
      baseURL: "https://ai.gitee.com/v1",
@ -48,27 +49,6 @@ class GiteeAILLM {
    console.log(`\x1b[36m[${this.constructor.name}]\x1b[0m ${text}`, ...args);
  }

-  // This checks if the .cached_at file has a timestamp that is more than 1Week (in millis)
-  // from the current date. If it is, then we will refetch the API so that all the models are up
-  // to date.
-  #cacheIsStale() {
-    const MAX_STALE = 6.048e8; // 1 Week in MS
-    if (!fs.existsSync(this.cacheAtPath)) return true;
-    const now = Number(new Date());
-    const timestampMs = Number(fs.readFileSync(this.cacheAtPath));
-    return now - timestampMs > MAX_STALE;
-  }
-
-  // This function fetches the models from the GiteeAI API and caches them locally.
-  async #syncModels() {
-    if (fs.existsSync(this.cacheModelPath) && !this.#cacheIsStale())
-      return false;
-
-    this.log("Model cache is not present or stale. Fetching from GiteeAI API.");
-    await giteeAiModels();
-    return;
-  }
-
  models() {
    if (!fs.existsSync(this.cacheModelPath)) return {};
    return safeJsonParse(
@ -171,6 +151,7 @@ class GiteeAILLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -187,6 +168,7 @@ class GiteeAILLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/groq/index.js
+++ b/server/utils/AiProviders/groq/index.js
@ -11,6 +11,7 @@ class GroqLLM {
  constructor(embedder = null, modelPreference = null) {
    const { OpenAI: OpenAIApi } = require("openai");
    if (!process.env.GROQ_API_KEY) throw new Error("No Groq API key was set.");
+    this.className = "GroqLLM";

    this.openai = new OpenAIApi({
      baseURL: "https://api.groq.com/openai/v1",
@ -204,6 +205,7 @@ class GroqLLM {
          result.output.usage.completion_time,
        duration: result.output.usage.total_time,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -225,6 +227,7 @@ class GroqLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/huggingface/index.js
+++ b/server/utils/AiProviders/huggingface/index.js
@ -14,6 +14,7 @@ class HuggingFaceLLM {
      throw new Error("No HuggingFace Access Token was set.");
    const { OpenAI: OpenAIApi } = require("openai");

+    this.className = "HuggingFaceLLM";
    this.openai = new OpenAIApi({
      baseURL: `${process.env.HUGGING_FACE_LLM_ENDPOINT}/v1`,
      apiKey: process.env.HUGGING_FACE_LLM_API_KEY,
@ -118,6 +119,7 @@ class HuggingFaceLLM {
          (result.output.usage?.completion_tokens || 0) / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -134,6 +136,7 @@ class HuggingFaceLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/koboldCPP/index.js
+++ b/server/utils/AiProviders/koboldCPP/index.js
@ -161,6 +161,7 @@ class KoboldCPPLLM {
        outputTps: completionTokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -178,6 +179,7 @@ class KoboldCPPLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/liteLLM/index.js
+++ b/server/utils/AiProviders/liteLLM/index.js
@ -155,6 +155,7 @@ class LiteLLM {
          (result.output.usage?.completion_tokens || 0) / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -170,9 +171,9 @@ class LiteLLM {
        max_tokens: parseInt(this.maxTokens), // LiteLLM requires int
      }),
      messages,
-      // runPromptTokenCalculation: true - We manually count the tokens because they may or may not be provided in the stream
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/lmStudio/index.js
+++ b/server/utils/AiProviders/lmStudio/index.js
@ -17,6 +17,7 @@ class LMStudioLLM {
    if (!process.env.LMSTUDIO_BASE_PATH)
      throw new Error("No LMStudio API Base Path was set.");

+    this.className = "LMStudioLLM";
    const apiKey = process.env.LMSTUDIO_AUTH_TOKEN ?? null;
    this.lmstudio = new OpenAIApi({
      baseURL: parseLMStudioBasePath(process.env.LMSTUDIO_BASE_PATH), // here is the URL to your LMStudio instance
@ -240,6 +241,7 @@ class LMStudioLLM {
        outputTps: result.output.usage?.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -261,6 +263,7 @@ class LMStudioLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/localAi/index.js
+++ b/server/utils/AiProviders/localAi/index.js
@ -12,6 +12,7 @@ class LocalAiLLM {
    if (!process.env.LOCAL_AI_BASE_PATH)
      throw new Error("No LocalAI Base Path was set.");

+    this.className = "LocalAiLLM";
    const { OpenAI: OpenAIApi } = require("openai");
    this.openai = new OpenAIApi({
      baseURL: process.env.LOCAL_AI_BASE_PATH,
@ -146,6 +147,7 @@ class LocalAiLLM {
        outputTps: completionTokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -167,6 +169,7 @@ class LocalAiLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/mistral/index.js
+++ b/server/utils/AiProviders/mistral/index.js
@ -140,6 +140,7 @@ class MistralLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -161,6 +162,7 @@ class MistralLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/moonshotAi/index.js
+++ b/server/utils/AiProviders/moonshotAi/index.js
@ -137,6 +137,7 @@ class MoonshotAiLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -153,6 +154,7 @@ class MoonshotAiLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/novita/index.js
+++ b/server/utils/AiProviders/novita/index.js
@ -226,6 +226,7 @@ class NovitaLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -247,6 +248,7 @@ class NovitaLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/nvidiaNim/index.js
+++ b/server/utils/AiProviders/nvidiaNim/index.js
@ -185,6 +185,7 @@ class NvidiaNimLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -206,6 +207,7 @@ class NvidiaNimLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/ollama/index.js
+++ b/server/utils/AiProviders/ollama/index.js
@ -312,6 +312,7 @@ class OllamaAILLM {
          result.output.usage.completion_tokens / result.output.usage.duration,
        duration: result.output.usage.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -333,6 +334,7 @@ class OllamaAILLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    }).catch((e) => {
      throw this.#errorHandler(e);
    });
--- a/server/utils/AiProviders/openAi/index.js
+++ b/server/utils/AiProviders/openAi/index.js
@ -176,6 +176,7 @@ class OpenAiLLM {
          : 0,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -198,6 +199,7 @@ class OpenAiLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/openRouter/index.js
+++ b/server/utils/AiProviders/openRouter/index.js
@ -277,6 +277,7 @@ class OpenRouterLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -311,6 +312,7 @@ class OpenRouterLLM {
      // is not available.
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/perplexity/index.js
+++ b/server/utils/AiProviders/perplexity/index.js
@ -18,6 +18,7 @@ class PerplexityLLM {
    if (!process.env.PERPLEXITY_API_KEY)
      throw new Error("No Perplexity API key was set.");

+    this.className = "PerplexityLLM";
    const { OpenAI: OpenAIApi } = require("openai");
    this.openai = new OpenAIApi({
      baseURL: "https://api.perplexity.ai",
@ -118,6 +119,7 @@ class PerplexityLLM {
        outputTps: result.output.usage?.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -139,6 +141,7 @@ class PerplexityLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/ppio/index.js
+++ b/server/utils/AiProviders/ppio/index.js
@ -177,6 +177,7 @@ class PPIOLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -198,6 +199,7 @@ class PPIOLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/privatemode/index.js
+++ b/server/utils/AiProviders/privatemode/index.js
@ -169,6 +169,7 @@ class PrivatemodeLLM {
        outputTps: result.output.usage?.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -190,6 +191,7 @@ class PrivatemodeLLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/sambanova/index.js
+++ b/server/utils/AiProviders/sambanova/index.js
@ -137,6 +137,7 @@ class SambaNovaLLM {
        outputTps: result.output.usage?.total_tokens_per_sec || 0,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -156,6 +157,7 @@ class SambaNovaLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/textGenWebUI/index.js
+++ b/server/utils/AiProviders/textGenWebUI/index.js
@ -151,6 +151,7 @@ class TextGenWebUILLM {
        outputTps: result.output.usage?.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -167,6 +168,7 @@ class TextGenWebUILLM {
      messages,
      runPromptTokenCalculation: true,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/togetherAi/index.js
+++ b/server/utils/AiProviders/togetherAi/index.js
@ -82,6 +82,7 @@ class TogetherAiLLM {
    if (!process.env.TOGETHER_AI_API_KEY)
      throw new Error("No TogetherAI API key was set.");
    const { OpenAI: OpenAIApi } = require("openai");
+    this.className = "TogetherAiLLM";
    this.openai = new OpenAIApi({
      baseURL: "https://api.together.xyz/v1",
      apiKey: process.env.TOGETHER_AI_API_KEY ?? null,
@ -210,6 +211,7 @@ class TogetherAiLLM {
        outputTps: result.output.usage?.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -231,6 +233,7 @@ class TogetherAiLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });
    return measuredStreamRequest;
  }
--- a/server/utils/AiProviders/xai/index.js
+++ b/server/utils/AiProviders/xai/index.js
@ -148,6 +148,7 @@ class XAiLLM {
        outputTps: result.output.usage.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -169,6 +170,7 @@ class XAiLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/AiProviders/zai/index.js
+++ b/server/utils/AiProviders/zai/index.js
@ -137,6 +137,7 @@ class ZAiLLM {
        outputTps: result.output.usage?.completion_tokens / result.duration,
        duration: result.duration,
        model: this.model,
+        provider: this.className,
        timestamp: new Date(),
      },
    };
@ -153,6 +154,7 @@ class ZAiLLM {
      messages,
      runPromptTokenCalculation: false,
      modelTag: this.model,
+      provider: this.className,
    });

    return measuredStreamRequest;
--- a/server/utils/helpers/chat/LLMPerformanceMonitor.js
+++ b/server/utils/helpers/chat/LLMPerformanceMonitor.js
@ -62,6 +62,7 @@ class LLMPerformanceMonitor {
   * @param {Messages} [opts.messages=[]] - the messages sent to the LLM so we can calculate the prompt tokens since most providers do not return this on stream
   * @param {boolean} [opts.runPromptTokenCalculation=true] - whether to run the prompt token calculation to estimate the `prompt_tokens` metric. This is useful for providers that do not return this on stream.
   * @param {string} [opts.modelTag=""] - the tag of the model that was used to generate the stream (eg: gpt-4o, claude-3-5-sonnet, qwen3/72b-instruct, etc.)
+   * @param {string} [opts.provider=""] - the class name of the LLM that was used to generate the stream (eg: OpenAI, Anthropic, LMStudio, ApiPie, etc.)
   * @returns {Promise<MonitoredStream>}
   */
  static async measureStream({
@ -69,6 +70,7 @@ class LLMPerformanceMonitor {
    messages = [],
    runPromptTokenCalculation = true,
    modelTag = "",
+    provider = "",
  }) {
    const stream = await func;
    stream.start = Date.now();
@ -80,6 +82,7 @@ class LLMPerformanceMonitor {
      outputTps: 0,
      duration: 0,
      ...(modelTag ? { model: modelTag } : {}),
+      ...(provider ? { provider: provider } : {}),
    };

    stream.endMeasurement = (reportedUsage = {}) => {