sort grouped models by org

update tagline
update comment

README.md

@@ -72,6 +72,7 @@ Some cool features of AnythingLLM
 - [LocalAi (all models)](https://localai.io/)
 - [Together AI (chat models)](https://www.together.ai/)
 - [Perplexity (chat models)](https://www.perplexity.ai/)
+- [OpenRouter (chat models)](https://openrouter.ai/)
 - [Mistral](https://mistral.ai/)
 
 **Supported Embedding models:**

frontend/src/components/LLMSelection/OpenRouterOptions/index.jsx

@@ -25,20 +25,29 @@ export default function OpenRouterOptions({ settings }) {
 }
 
 function OpenRouterModelSelection({ settings }) {
-  const [customModels, setCustomModels] = useState([]);
+  const [groupedModels, setGroupedModels] = useState({});
   const [loading, setLoading] = useState(true);
 
   useEffect(() => {
     async function findCustomModels() {
       setLoading(true);
       const { models } = await System.customModels("openrouter");
-      setCustomModels(models || []);
+      if (models?.length > 0) {
+        const modelsByOrganization = models.reduce((acc, model) => {
+          acc[model.organization] = acc[model.organization] || [];
+          acc[model.organization].push(model);
+          return acc;
+        }, {});
+
+        setGroupedModels(modelsByOrganization);
+      }
+
       setLoading(false);
     }
     findCustomModels();
   }, []);
 
-  if (loading || customModels.length == 0) {
+  if (loading || Object.keys(groupedModels).length === 0) {
     return (
       <div className="flex flex-col w-60">
         <label className="text-white text-sm font-semibold block mb-4">
@@ -67,21 +76,21 @@ function OpenRouterModelSelection({ settings }) {
         required={true}
         className="bg-zinc-900 border border-gray-500 text-white text-sm rounded-lg block w-full p-2.5"
       >
-        {customModels.length > 0 && (
-          <optgroup label="Available OpenRouter Models">
-            {customModels.map((model) => {
-              return (
+        {Object.keys(groupedModels)
+          .sort()
+          .map((organization) => (
+            <optgroup key={organization} label={organization}>
+              {groupedModels[organization].map((model) => (
                 <option
                   key={model.id}
                   value={model.id}
-                  selected={settings?.OpenRouterModelPref === model.id}
+                  selected={settings.OpenRouterModelPref === model.id}
                 >
-                  {model.id}
+                  {model.name}
                 </option>
-              );
-            })}
-          </optgroup>
-        )}
+              ))}
+            </optgroup>
+          ))}
       </select>
     </div>
   );

frontend/src/components/LLMSelection/TogetherAiOptions/index.jsx

@@ -76,19 +76,21 @@ function TogetherAiModelSelection({ settings }) {
         required={true}
         className="bg-zinc-900 border border-gray-500 text-white text-sm rounded-lg block w-full p-2.5"
       >
-        {Object.entries(groupedModels).map(([organization, models]) => (
-          <optgroup key={organization} label={organization}>
-            {models.map((model) => (
-              <option
-                key={model.id}
-                value={model.id}
-                selected={settings.TogetherAiModelPref === model.id}
-              >
-                {model.name}
-              </option>
-            ))}
-          </optgroup>
-        ))}
+        {Object.keys(groupedModels)
+          .sort()
+          .map((organization) => (
+            <optgroup key={organization} label={organization}>
+              {groupedModels[organization].map((model) => (
+                <option
+                  key={model.id}
+                  value={model.id}
+                  selected={settings.OpenRouterModelPref === model.id}
+                >
+                  {model.name}
+                </option>
+              ))}
+            </optgroup>
+          ))}
       </select>
     </div>
   );

frontend/src/pages/GeneralSettings/LLMPreference/index.jsx

@@ -171,7 +171,7 @@ export default function GeneralLLMPreference() {
       value: "openrouter",
       logo: OpenRouterLogo,
       options: <OpenRouterOptions settings={settings} />,
-      description: "Run any model from OpenRouter.",
+      description: "A unified interface for LLMs.",
     },
     {
       name: "Native",

frontend/src/pages/OnboardingFlow/Steps/DataHandling/index.jsx

@@ -122,6 +122,7 @@ const LLM_SELECTION_PRIVACY = {
   openrouter: {
     name: "OpenRouter",
     description: [
+      "Your chats will not be used for training",
       "Your prompts and document text used in response creation are visible to OpenRouter",
     ],
     logo: OpenRouterLogo,

frontend/src/pages/OnboardingFlow/Steps/LLMPreference/index.jsx

@@ -145,7 +145,7 @@ export default function LLMPreference({
       value: "openrouter",
       logo: OpenRouterLogo,
       options: <OpenRouterOptions settings={settings} />,
-      description: "Run any model from OpenRouter.",
+      description: "A unified interface for LLMs.",
     },
     {
       name: "Native",

server/utils/AiProviders/openRouter/index.js

@@ -1,24 +1,11 @@
 const { NativeEmbedder } = require("../../EmbeddingEngines/native");
 const { chatPrompt } = require("../../chats");
-const { handleDefaultStreamResponse } = require("../../helpers/chat/responses");
+const { v4: uuidv4 } = require("uuid");
+const { writeResponseChunk } = require("../../helpers/chat/responses");
 
-async function openRouterModels() {
-  try {
-    const response = await fetch("https://openrouter.ai/api/v1/models");
-    const data = await response.json();
-    let MODELS = {};
-    data.data.forEach((model) => {
-      MODELS[model.id] = {
-        id: model.id,
-        name: model.name,
-        maxLength: model.context_length,
-      };
-    });
-    return MODELS;
-  } catch (e) {
-    console.error(e);
-    return {};
-  }
+function openRouterModels() {
+  const { MODELS } = require("./models.js");
+  return MODELS || {};
 }
 
 class OpenRouterLLM {
@@ -30,10 +17,11 @@ class OpenRouterLLM {
     const config = new Configuration({
       basePath: "https://openrouter.ai/api/v1",
       apiKey: process.env.OPENROUTER_API_KEY,
-      // TODO: not working to update usage dashboard
-      defaultHeaders: {
-        "HTTP-Referer": "https://useanything.com",
-        "X-Title": "AnythingLLM",
+      baseOptions: {
+        headers: {
+          "HTTP-Referer": "https://useanything.com",
+          "X-Title": "AnythingLLM",
+        },
       },
     });
     this.openai = new OpenAIApi(config);
@@ -61,8 +49,8 @@ class OpenRouterLLM {
     );
   }
 
-  async allModelInformation() {
-    return await openRouterModels();
+  allModelInformation() {
+    return openRouterModels();
   }
 
   streamingEnabled() {
@@ -70,15 +58,13 @@ class OpenRouterLLM {
   }
 
   promptWindowLimit() {
-    // TODO: get from openRouterModels()
-    // const availableModels = this.allModelInformation();
-    // return availableModels[this.model]?.maxLength || 4096;
-    return 4096;
+    const availableModels = this.allModelInformation();
+    return availableModels[this.model]?.maxLength || 4096;
   }
 
   async isValidChatCompletionModel(model = "") {
-    // TODO: get from openRouterModels()
-    return true;
+    const availableModels = this.allModelInformation();
+    return availableModels.hasOwnProperty(model);
   }
 
   constructPrompt({
@@ -201,7 +187,130 @@ class OpenRouterLLM {
   }
 
   handleStream(response, stream, responseProps) {
-    return handleDefaultStreamResponse(response, stream, responseProps);
+    const timeoutThresholdMs = 500;
+    const { uuid = uuidv4(), sources = [] } = responseProps;
+
+    return new Promise((resolve) => {
+      let fullText = "";
+      let chunk = "";
+      let lastChunkTime = null; // null when first token is still not received.
+
+      // NOTICE: Not all OpenRouter models will return a stop reason
+      // which keeps the connection open and so the model never finalizes the stream
+      // like the traditional OpenAI response schema does. So in the case the response stream
+      // never reaches a formal close state we maintain an interval timer that if we go >=timeoutThresholdMs with
+      // no new chunks then we kill the stream and assume it to be complete. OpenRouter is quite fast
+      // so this threshold should permit most responses, but we can adjust `timeoutThresholdMs` if
+      // we find it is too aggressive.
+      const timeoutCheck = setInterval(() => {
+        if (lastChunkTime === null) return;
+
+        const now = Number(new Date());
+        const diffMs = now - lastChunkTime;
+        if (diffMs >= timeoutThresholdMs) {
+          console.log(
+            `OpenRouter stream did not self-close and has been stale for >${timeoutThresholdMs}ms. Closing response stream.`
+          );
+          writeResponseChunk(response, {
+            uuid,
+            sources,
+            type: "textResponseChunk",
+            textResponse: "",
+            close: true,
+            error: false,
+          });
+          clearInterval(timeoutCheck);
+          resolve(fullText);
+        }
+      }, 500);
+
+      stream.data.on("data", (data) => {
+        const lines = data
+          ?.toString()
+          ?.split("\n")
+          .filter((line) => line.trim() !== "");
+
+        for (const line of lines) {
+          let validJSON = false;
+          const message = chunk + line.replace(/^data: /, "");
+
+          // JSON chunk is incomplete and has not ended yet
+          // so we need to stitch it together. You would think JSON
+          // chunks would only come complete - but they don't!
+          try {
+            JSON.parse(message);
+            validJSON = true;
+          } catch {}
+
+          if (!validJSON) {
+            // It can be possible that the chunk decoding is running away
+            // and the message chunk fails to append due to string length.
+            // In this case abort the chunk and reset so we can continue.
+            // ref: https://github.com/Mintplex-Labs/anything-llm/issues/416
+            try {
+              chunk += message;
+            } catch (e) {
+              console.error(`Chunk appending error`, e);
+              chunk = "";
+            }
+            continue;
+          } else {
+            chunk = "";
+          }
+
+          if (message == "[DONE]") {
+            lastChunkTime = Number(new Date());
+            writeResponseChunk(response, {
+              uuid,
+              sources,
+              type: "textResponseChunk",
+              textResponse: "",
+              close: true,
+              error: false,
+            });
+            clearInterval(timeoutCheck);
+            resolve(fullText);
+          } else {
+            let finishReason = null;
+            let token = "";
+            try {
+              const json = JSON.parse(message);
+              token = json?.choices?.[0]?.delta?.content;
+              finishReason = json?.choices?.[0]?.finish_reason || null;
+            } catch {
+              continue;
+            }
+
+            if (token) {
+              fullText += token;
+              lastChunkTime = Number(new Date());
+              writeResponseChunk(response, {
+                uuid,
+                sources: [],
+                type: "textResponseChunk",
+                textResponse: token,
+                close: false,
+                error: false,
+              });
+            }
+
+            if (finishReason !== null) {
+              lastChunkTime = Number(new Date());
+              writeResponseChunk(response, {
+                uuid,
+                sources,
+                type: "textResponseChunk",
+                textResponse: "",
+                close: true,
+                error: false,
+              });
+              clearInterval(timeoutCheck);
+              resolve(fullText);
+            }
+          }
+        }
+      });
+    });
   }
 
   // Simple wrapper for dynamic embedder & normalize interface for all LLM implementations

server/utils/AiProviders/openRouter/models.js

@@ -0,0 +1,622 @@
+const MODELS = {
+  "nousresearch/nous-capybara-34b": {
+    id: "nousresearch/nous-capybara-34b",
+    name: "Nous: Capybara 34B",
+    organization: "Nousresearch",
+    maxLength: 32768,
+  },
+  "openrouter/auto": {
+    id: "openrouter/auto",
+    name: "Auto (best for prompt)",
+    organization: "Openrouter",
+    maxLength: 128000,
+  },
+  "nousresearch/nous-capybara-7b:free": {
+    id: "nousresearch/nous-capybara-7b:free",
+    name: "Nous: Capybara 7B (free)",
+    organization: "Nousresearch",
+    maxLength: 4096,
+  },
+  "mistralai/mistral-7b-instruct:free": {
+    id: "mistralai/mistral-7b-instruct:free",
+    name: "Mistral 7B Instruct (free)",
+    organization: "Mistralai",
+    maxLength: 8192,
+  },
+  "gryphe/mythomist-7b:free": {
+    id: "gryphe/mythomist-7b:free",
+    name: "MythoMist 7B (free)",
+    organization: "Gryphe",
+    maxLength: 32768,
+  },
+  "undi95/toppy-m-7b:free": {
+    id: "undi95/toppy-m-7b:free",
+    name: "Toppy M 7B (free)",
+    organization: "Undi95",
+    maxLength: 4096,
+  },
+  "openrouter/cinematika-7b:free": {
+    id: "openrouter/cinematika-7b:free",
+    name: "Cinematika 7B (alpha) (free)",
+    organization: "Openrouter",
+    maxLength: 8000,
+  },
+  "google/gemma-7b-it:free": {
+    id: "google/gemma-7b-it:free",
+    name: "Google: Gemma 7B (free)",
+    organization: "Google",
+    maxLength: 8000,
+  },
+  "jondurbin/bagel-34b": {
+    id: "jondurbin/bagel-34b",
+    name: "Bagel 34B v0.2",
+    organization: "Jondurbin",
+    maxLength: 8000,
+  },
+  "jebcarter/psyfighter-13b": {
+    id: "jebcarter/psyfighter-13b",
+    name: "Psyfighter 13B",
+    organization: "Jebcarter",
+    maxLength: 4096,
+  },
+  "koboldai/psyfighter-13b-2": {
+    id: "koboldai/psyfighter-13b-2",
+    name: "Psyfighter v2 13B",
+    organization: "Koboldai",
+    maxLength: 4096,
+  },
+  "neversleep/noromaid-mixtral-8x7b-instruct": {
+    id: "neversleep/noromaid-mixtral-8x7b-instruct",
+    name: "Noromaid Mixtral 8x7B Instruct",
+    organization: "Neversleep",
+    maxLength: 8000,
+  },
+  "nousresearch/nous-hermes-llama2-13b": {
+    id: "nousresearch/nous-hermes-llama2-13b",
+    name: "Nous: Hermes 13B",
+    organization: "Nousresearch",
+    maxLength: 4096,
+  },
+  "meta-llama/codellama-34b-instruct": {
+    id: "meta-llama/codellama-34b-instruct",
+    name: "Meta: CodeLlama 34B Instruct",
+    organization: "Meta-llama",
+    maxLength: 8192,
+  },
+  "phind/phind-codellama-34b": {
+    id: "phind/phind-codellama-34b",
+    name: "Phind: CodeLlama 34B v2",
+    organization: "Phind",
+    maxLength: 4096,
+  },
+  "intel/neural-chat-7b": {
+    id: "intel/neural-chat-7b",
+    name: "Neural Chat 7B v3.1",
+    organization: "Intel",
+    maxLength: 4096,
+  },
+  "mistralai/mixtral-8x7b-instruct": {
+    id: "mistralai/mixtral-8x7b-instruct",
+    name: "Mistral: Mixtral 8x7B Instruct",
+    organization: "Mistralai",
+    maxLength: 32768,
+  },
+  "nousresearch/nous-hermes-2-mixtral-8x7b-dpo": {
+    id: "nousresearch/nous-hermes-2-mixtral-8x7b-dpo",
+    name: "Nous: Hermes 2 Mixtral 8x7B DPO",
+    organization: "Nousresearch",
+    maxLength: 32000,
+  },
+  "nousresearch/nous-hermes-2-mixtral-8x7b-sft": {
+    id: "nousresearch/nous-hermes-2-mixtral-8x7b-sft",
+    name: "Nous: Hermes 2 Mixtral 8x7B SFT",
+    organization: "Nousresearch",
+    maxLength: 32000,
+  },
+  "haotian-liu/llava-13b": {
+    id: "haotian-liu/llava-13b",
+    name: "Llava 13B",
+    organization: "Haotian-liu",
+    maxLength: 2048,
+  },
+  "nousresearch/nous-hermes-2-vision-7b": {
+    id: "nousresearch/nous-hermes-2-vision-7b",
+    name: "Nous: Hermes 2 Vision 7B (alpha)",
+    organization: "Nousresearch",
+    maxLength: 4096,
+  },
+  "meta-llama/llama-2-13b-chat": {
+    id: "meta-llama/llama-2-13b-chat",
+    name: "Meta: Llama v2 13B Chat",
+    organization: "Meta-llama",
+    maxLength: 4096,
+  },
+  "migtissera/synthia-70b": {
+    id: "migtissera/synthia-70b",
+    name: "Synthia 70B",
+    organization: "Migtissera",
+    maxLength: 8192,
+  },
+  "pygmalionai/mythalion-13b": {
+    id: "pygmalionai/mythalion-13b",
+    name: "Pygmalion: Mythalion 13B",
+    organization: "Pygmalionai",
+    maxLength: 8192,
+  },
+  "undi95/remm-slerp-l2-13b-6k": {
+    id: "undi95/remm-slerp-l2-13b-6k",
+    name: "ReMM SLERP 13B 6k",
+    organization: "Undi95",
+    maxLength: 6144,
+  },
+  "gryphe/mythomax-l2-13b": {
+    id: "gryphe/mythomax-l2-13b",
+    name: "MythoMax 13B",
+    organization: "Gryphe",
+    maxLength: 4096,
+  },
+  "xwin-lm/xwin-lm-70b": {
+    id: "xwin-lm/xwin-lm-70b",
+    name: "Xwin 70B",
+    organization: "Xwin-lm",
+    maxLength: 8192,
+  },
+  "gryphe/mythomax-l2-13b-8k": {
+    id: "gryphe/mythomax-l2-13b-8k",
+    name: "MythoMax 13B 8k",
+    organization: "Gryphe",
+    maxLength: 8192,
+  },
+  "alpindale/goliath-120b": {
+    id: "alpindale/goliath-120b",
+    name: "Goliath 120B",
+    organization: "Alpindale",
+    maxLength: 6144,
+  },
+  "neversleep/noromaid-20b": {
+    id: "neversleep/noromaid-20b",
+    name: "Noromaid 20B",
+    organization: "Neversleep",
+    maxLength: 8192,
+  },
+  "gryphe/mythomist-7b": {
+    id: "gryphe/mythomist-7b",
+    name: "MythoMist 7B",
+    organization: "Gryphe",
+    maxLength: 32768,
+  },
+  "mancer/weaver": {
+    id: "mancer/weaver",
+    name: "Mancer: Weaver (alpha)",
+    organization: "Mancer",
+    maxLength: 8000,
+  },
+  "nousresearch/nous-hermes-llama2-70b": {
+    id: "nousresearch/nous-hermes-llama2-70b",
+    name: "Nous: Hermes 70B",
+    organization: "Nousresearch",
+    maxLength: 4096,
+  },
+  "nousresearch/nous-capybara-7b": {
+    id: "nousresearch/nous-capybara-7b",
+    name: "Nous: Capybara 7B",
+    organization: "Nousresearch",
+    maxLength: 4096,
+  },
+  "codellama/codellama-70b-instruct": {
+    id: "codellama/codellama-70b-instruct",
+    name: "Meta: CodeLlama 70B Instruct",
+    organization: "Codellama",
+    maxLength: 2048,
+  },
+  "teknium/openhermes-2-mistral-7b": {
+    id: "teknium/openhermes-2-mistral-7b",
+    name: "OpenHermes 2 Mistral 7B",
+    organization: "Teknium",
+    maxLength: 4096,
+  },
+  "teknium/openhermes-2.5-mistral-7b": {
+    id: "teknium/openhermes-2.5-mistral-7b",
+    name: "OpenHermes 2.5 Mistral 7B",
+    organization: "Teknium",
+    maxLength: 4096,
+  },
+  "undi95/remm-slerp-l2-13b": {
+    id: "undi95/remm-slerp-l2-13b",
+    name: "ReMM SLERP 13B",
+    organization: "Undi95",
+    maxLength: 4096,
+  },
+  "undi95/toppy-m-7b": {
+    id: "undi95/toppy-m-7b",
+    name: "Toppy M 7B",
+    organization: "Undi95",
+    maxLength: 4096,
+  },
+  "openrouter/cinematika-7b": {
+    id: "openrouter/cinematika-7b",
+    name: "Cinematika 7B (alpha)",
+    organization: "Openrouter",
+    maxLength: 8000,
+  },
+  "01-ai/yi-34b-chat": {
+    id: "01-ai/yi-34b-chat",
+    name: "Yi 34B Chat",
+    organization: "01-ai",
+    maxLength: 4096,
+  },
+  "01-ai/yi-34b": {
+    id: "01-ai/yi-34b",
+    name: "Yi 34B (base)",
+    organization: "01-ai",
+    maxLength: 4096,
+  },
+  "01-ai/yi-6b": {
+    id: "01-ai/yi-6b",
+    name: "Yi 6B (base)",
+    organization: "01-ai",
+    maxLength: 4096,
+  },
+  "togethercomputer/stripedhyena-nous-7b": {
+    id: "togethercomputer/stripedhyena-nous-7b",
+    name: "StripedHyena Nous 7B",
+    organization: "Togethercomputer",
+    maxLength: 32768,
+  },
+  "togethercomputer/stripedhyena-hessian-7b": {
+    id: "togethercomputer/stripedhyena-hessian-7b",
+    name: "StripedHyena Hessian 7B (base)",
+    organization: "Togethercomputer",
+    maxLength: 32768,
+  },
+  "mistralai/mixtral-8x7b": {
+    id: "mistralai/mixtral-8x7b",
+    name: "Mistral: Mixtral 8x7B (base)",
+    organization: "Mistralai",
+    maxLength: 32768,
+  },
+  "nousresearch/nous-hermes-yi-34b": {
+    id: "nousresearch/nous-hermes-yi-34b",
+    name: "Nous: Hermes 2 Yi 34B",
+    organization: "Nousresearch",
+    maxLength: 4096,
+  },
+  "nousresearch/nous-hermes-2-mistral-7b-dpo": {
+    id: "nousresearch/nous-hermes-2-mistral-7b-dpo",
+    name: "Nous: Hermes 2 Mistral 7B DPO",
+    organization: "Nousresearch",
+    maxLength: 8192,
+  },
+  "open-orca/mistral-7b-openorca": {
+    id: "open-orca/mistral-7b-openorca",
+    name: "Mistral OpenOrca 7B",
+    organization: "Open-orca",
+    maxLength: 8192,
+  },
+  "huggingfaceh4/zephyr-7b-beta": {
+    id: "huggingfaceh4/zephyr-7b-beta",
+    name: "Hugging Face: Zephyr 7B",
+    organization: "Huggingfaceh4",
+    maxLength: 4096,
+  },
+  "openai/gpt-3.5-turbo": {
+    id: "openai/gpt-3.5-turbo",
+    name: "OpenAI: GPT-3.5 Turbo",
+    organization: "Openai",
+    maxLength: 4095,
+  },
+  "openai/gpt-3.5-turbo-0125": {
+    id: "openai/gpt-3.5-turbo-0125",
+    name: "OpenAI: GPT-3.5 Turbo 16k",
+    organization: "Openai",
+    maxLength: 16385,
+  },
+  "openai/gpt-3.5-turbo-1106": {
+    id: "openai/gpt-3.5-turbo-1106",
+    name: "OpenAI: GPT-3.5 Turbo 16k (older v1106)",
+    organization: "Openai",
+    maxLength: 16385,
+  },
+  "openai/gpt-3.5-turbo-0613": {
+    id: "openai/gpt-3.5-turbo-0613",
+    name: "OpenAI: GPT-3.5 Turbo (older v0613)",
+    organization: "Openai",
+    maxLength: 4095,
+  },
+  "openai/gpt-3.5-turbo-0301": {
+    id: "openai/gpt-3.5-turbo-0301",
+    name: "OpenAI: GPT-3.5 Turbo (older v0301)",
+    organization: "Openai",
+    maxLength: 4095,
+  },
+  "openai/gpt-3.5-turbo-16k": {
+    id: "openai/gpt-3.5-turbo-16k",
+    name: "OpenAI: GPT-3.5 Turbo 16k",
+    organization: "Openai",
+    maxLength: 16385,
+  },
+  "openai/gpt-4-turbo-preview": {
+    id: "openai/gpt-4-turbo-preview",
+    name: "OpenAI: GPT-4 Turbo (preview)",
+    organization: "Openai",
+    maxLength: 128000,
+  },
+  "openai/gpt-4-1106-preview": {
+    id: "openai/gpt-4-1106-preview",
+    name: "OpenAI: GPT-4 Turbo (older v1106)",
+    organization: "Openai",
+    maxLength: 128000,
+  },
+  "openai/gpt-4": {
+    id: "openai/gpt-4",
+    name: "OpenAI: GPT-4",
+    organization: "Openai",
+    maxLength: 8191,
+  },
+  "openai/gpt-4-0314": {
+    id: "openai/gpt-4-0314",
+    name: "OpenAI: GPT-4 (older v0314)",
+    organization: "Openai",
+    maxLength: 8191,
+  },
+  "openai/gpt-4-32k": {
+    id: "openai/gpt-4-32k",
+    name: "OpenAI: GPT-4 32k",
+    organization: "Openai",
+    maxLength: 32767,
+  },
+  "openai/gpt-4-32k-0314": {
+    id: "openai/gpt-4-32k-0314",
+    name: "OpenAI: GPT-4 32k (older v0314)",
+    organization: "Openai",
+    maxLength: 32767,
+  },
+  "openai/gpt-4-vision-preview": {
+    id: "openai/gpt-4-vision-preview",
+    name: "OpenAI: GPT-4 Vision (preview)",
+    organization: "Openai",
+    maxLength: 128000,
+  },
+  "openai/gpt-3.5-turbo-instruct": {
+    id: "openai/gpt-3.5-turbo-instruct",
+    name: "OpenAI: GPT-3.5 Turbo Instruct",
+    organization: "Openai",
+    maxLength: 4095,
+  },
+  "google/palm-2-chat-bison": {
+    id: "google/palm-2-chat-bison",
+    name: "Google: PaLM 2 Chat",
+    organization: "Google",
+    maxLength: 36864,
+  },
+  "google/palm-2-codechat-bison": {
+    id: "google/palm-2-codechat-bison",
+    name: "Google: PaLM 2 Code Chat",
+    organization: "Google",
+    maxLength: 28672,
+  },
+  "google/palm-2-chat-bison-32k": {
+    id: "google/palm-2-chat-bison-32k",
+    name: "Google: PaLM 2 Chat 32k",
+    organization: "Google",
+    maxLength: 131072,
+  },
+  "google/palm-2-codechat-bison-32k": {
+    id: "google/palm-2-codechat-bison-32k",
+    name: "Google: PaLM 2 Code Chat 32k",
+    organization: "Google",
+    maxLength: 131072,
+  },
+  "google/gemini-pro": {
+    id: "google/gemini-pro",
+    name: "Google: Gemini Pro (preview)",
+    organization: "Google",
+    maxLength: 131040,
+  },
+  "google/gemini-pro-vision": {
+    id: "google/gemini-pro-vision",
+    name: "Google: Gemini Pro Vision (preview)",
+    organization: "Google",
+    maxLength: 65536,
+  },
+  "perplexity/pplx-70b-online": {
+    id: "perplexity/pplx-70b-online",
+    name: "Perplexity: PPLX 70B Online",
+    organization: "Perplexity",
+    maxLength: 4096,
+  },
+  "perplexity/pplx-7b-online": {
+    id: "perplexity/pplx-7b-online",
+    name: "Perplexity: PPLX 7B Online",
+    organization: "Perplexity",
+    maxLength: 4096,
+  },
+  "perplexity/pplx-7b-chat": {
+    id: "perplexity/pplx-7b-chat",
+    name: "Perplexity: PPLX 7B Chat",
+    organization: "Perplexity",
+    maxLength: 8192,
+  },
+  "perplexity/pplx-70b-chat": {
+    id: "perplexity/pplx-70b-chat",
+    name: "Perplexity: PPLX 70B Chat",
+    organization: "Perplexity",
+    maxLength: 4096,
+  },
+  "meta-llama/llama-2-70b-chat": {
+    id: "meta-llama/llama-2-70b-chat",
+    name: "Meta: Llama v2 70B Chat",
+    organization: "Meta-llama",
+    maxLength: 4096,
+  },
+  "jondurbin/airoboros-l2-70b": {
+    id: "jondurbin/airoboros-l2-70b",
+    name: "Airoboros 70B",
+    organization: "Jondurbin",
+    maxLength: 4096,
+  },
+  "austism/chronos-hermes-13b": {
+    id: "austism/chronos-hermes-13b",
+    name: "Chronos Hermes 13B v2",
+    organization: "Austism",
+    maxLength: 4096,
+  },
+  "mistralai/mistral-7b-instruct": {
+    id: "mistralai/mistral-7b-instruct",
+    name: "Mistral 7B Instruct",
+    organization: "Mistralai",
+    maxLength: 8192,
+  },
+  "openchat/openchat-7b": {
+    id: "openchat/openchat-7b",
+    name: "OpenChat 3.5",
+    organization: "Openchat",
+    maxLength: 8192,
+  },
+  "lizpreciatior/lzlv-70b-fp16-hf": {
+    id: "lizpreciatior/lzlv-70b-fp16-hf",
+    name: "lzlv 70B",
+    organization: "Lizpreciatior",
+    maxLength: 4096,
+  },
+  "cognitivecomputations/dolphin-mixtral-8x7b": {
+    id: "cognitivecomputations/dolphin-mixtral-8x7b",
+    name: "Dolphin 2.6 Mixtral 8x7B 🐬",
+    organization: "Cognitivecomputations",
+    maxLength: 32000,
+  },
+  "rwkv/rwkv-5-world-3b": {
+    id: "rwkv/rwkv-5-world-3b",
+    name: "RWKV v5 World 3B",
+    organization: "Rwkv",
+    maxLength: 10000,
+  },
+  "recursal/rwkv-5-3b-ai-town": {
+    id: "recursal/rwkv-5-3b-ai-town",
+    name: "RWKV v5 3B AI Town",
+    organization: "Recursal",
+    maxLength: 10000,
+  },
+  "recursal/eagle-7b": {
+    id: "recursal/eagle-7b",
+    name: "RWKV v5: Eagle 7B",
+    organization: "Recursal",
+    maxLength: 10000,
+  },
+  "google/gemma-7b-it": {
+    id: "google/gemma-7b-it",
+    name: "Google: Gemma 7B",
+    organization: "Google",
+    maxLength: 8000,
+  },
+  "anthropic/claude-2": {
+    id: "anthropic/claude-2",
+    name: "Anthropic: Claude v2",
+    organization: "Anthropic",
+    maxLength: 200000,
+  },
+  "anthropic/claude-2.1": {
+    id: "anthropic/claude-2.1",
+    name: "Anthropic: Claude v2.1",
+    organization: "Anthropic",
+    maxLength: 200000,
+  },
+  "anthropic/claude-2.0": {
+    id: "anthropic/claude-2.0",
+    name: "Anthropic: Claude v2.0",
+    organization: "Anthropic",
+    maxLength: 100000,
+  },
+  "anthropic/claude-instant-1": {
+    id: "anthropic/claude-instant-1",
+    name: "Anthropic: Claude Instant v1",
+    organization: "Anthropic",
+    maxLength: 100000,
+  },
+  "anthropic/claude-instant-1.2": {
+    id: "anthropic/claude-instant-1.2",
+    name: "Anthropic: Claude Instant v1.2",
+    organization: "Anthropic",
+    maxLength: 100000,
+  },
+  "anthropic/claude-1": {
+    id: "anthropic/claude-1",
+    name: "Anthropic: Claude v1",
+    organization: "Anthropic",
+    maxLength: 100000,
+  },
+  "anthropic/claude-1.2": {
+    id: "anthropic/claude-1.2",
+    name: "Anthropic: Claude (older v1)",
+    organization: "Anthropic",
+    maxLength: 100000,
+  },
+  "anthropic/claude-instant-1.0": {
+    id: "anthropic/claude-instant-1.0",
+    name: "Anthropic: Claude Instant (older v1)",
+    organization: "Anthropic",
+    maxLength: 100000,
+  },
+  "anthropic/claude-instant-1.1": {
+    id: "anthropic/claude-instant-1.1",
+    name: "Anthropic: Claude Instant (older v1.1)",
+    organization: "Anthropic",
+    maxLength: 100000,
+  },
+  "anthropic/claude-2:beta": {
+    id: "anthropic/claude-2:beta",
+    name: "Anthropic: Claude v2 (experimental)",
+    organization: "Anthropic",
+    maxLength: 200000,
+  },
+  "anthropic/claude-2.1:beta": {
+    id: "anthropic/claude-2.1:beta",
+    name: "Anthropic: Claude v2.1 (experimental)",
+    organization: "Anthropic",
+    maxLength: 200000,
+  },
+  "anthropic/claude-2.0:beta": {
+    id: "anthropic/claude-2.0:beta",
+    name: "Anthropic: Claude v2.0 (experimental)",
+    organization: "Anthropic",
+    maxLength: 100000,
+  },
+  "anthropic/claude-instant-1:beta": {
+    id: "anthropic/claude-instant-1:beta",
+    name: "Anthropic: Claude Instant v1 (experimental)",
+    organization: "Anthropic",
+    maxLength: 100000,
+  },
+  "huggingfaceh4/zephyr-7b-beta:free": {
+    id: "huggingfaceh4/zephyr-7b-beta:free",
+    name: "Hugging Face: Zephyr 7B (free)",
+    organization: "Huggingfaceh4",
+    maxLength: 4096,
+  },
+  "openchat/openchat-7b:free": {
+    id: "openchat/openchat-7b:free",
+    name: "OpenChat 3.5 (free)",
+    organization: "Openchat",
+    maxLength: 8192,
+  },
+  "mistralai/mistral-tiny": {
+    id: "mistralai/mistral-tiny",
+    name: "Mistral: Tiny",
+    organization: "Mistralai",
+    maxLength: 32000,
+  },
+  "mistralai/mistral-small": {
+    id: "mistralai/mistral-small",
+    name: "Mistral: Small",
+    organization: "Mistralai",
+    maxLength: 32000,
+  },
+  "mistralai/mistral-medium": {
+    id: "mistralai/mistral-medium",
+    name: "Mistral: Medium",
+    organization: "Mistralai",
+    maxLength: 32000,
+  },
+};
+
+module.exports.MODELS = MODELS;

server/utils/AiProviders/openRouter/scripts/.gitignore

@@ -0,0 +1 @@
+*.json

server/utils/AiProviders/openRouter/scripts/parse.mjs

@@ -0,0 +1,37 @@
+// OpenRouter has lots of models we can use so we use this script
+// to cache all the models. We can see the list of all the models
+// here: https://openrouter.ai/docs#models
+
+// To run, cd into this directory and run `node parse.mjs`
+// copy outputs into the export in ../models.js
+
+// Update the date below if you run this again because OpenRouter added new models.
+// Last Collected: Feb 23, 2024
+
+import fs from "fs";
+
+async function parseChatModels() {
+  const models = {};
+  const response = await fetch("https://openrouter.ai/api/v1/models");
+  const data = await response.json();
+  data.data.forEach((model) => {
+    models[model.id] = {
+      id: model.id,
+      name: model.name,
+      // capitalize first letter
+      organization:
+        model.id.split("/")[0].charAt(0).toUpperCase() +
+        model.id.split("/")[0].slice(1),
+      maxLength: model.context_length,
+    };
+  });
+
+  fs.writeFileSync(
+    "chat_models.json",
+    JSON.stringify(models, null, 2),
+    "utf-8"
+  );
+  return models;
+}
+
+parseChatModels();

server/utils/helpers/customModels.js

@@ -150,6 +150,7 @@ async function getOpenRouterModels() {
   const models = Object.values(knownModels).map((model) => {
     return {
       id: model.id,
+      organization: model.organization,
       name: model.name,
     };
   });