@@ -90,7 +91,7 @@ AnythingLLM اسناد شما را به اشیایی به نام `workspaces` ت
- [LM Studio (all models)](https://lmstudio.ai)
- [LocalAi (all models)](https://localai.io/)
- [Together AI (chat models)](https://www.together.ai/)
-- [Fireworks AI (chat models)](https://fireworks.ai/)
+- [Fireworks AI (chat models)](https://fireworks.ai/)
- [Perplexity (chat models)](https://www.perplexity.ai/)
- [OpenRouter (chat models)](https://openrouter.ai/)
- [DeepSeek (chat models)](https://deepseek.com/)
@@ -106,6 +107,8 @@ AnythingLLM اسناد شما را به اشیایی به نام `workspaces` ت
- [Novita AI (chat models)](https://novita.ai/model-api/product/llm-api?utm_source=github_anything-llm&utm_medium=github_readme&utm_campaign=link)
- [PPIO](https://ppinfra.com?utm_source=github_anything-llm)
- [Docker Model Runner](https://docs.docker.com/ai/model-runner/)
+- [PrivateModeAI (chat models)](https://privatemode.ai/)
+- [SambaNova Cloud (chat models)](https://cloud.sambanova.ai/)
@@ -166,14 +169,15 @@ AnythingLLM اسناد شما را به اشیایی به نام `workspaces` ت
Mintplex Labs و جامعه کاربران، روشها، اسکریپتها و قالبهای متعددی را برای اجرای AnythingLLM به صورت محلی نگهداری میکنند. برای مطالعه نحوه استقرار در محیط مورد نظر خود یا استقرار خودکار، به جدول زیر مراجعه کنید.
+
-| Docker | AWS | GCP | Digital Ocean | Render.com |
-|----------------------------------------|----|-----|---------------|------------|
+| Docker | AWS | GCP | Digital Ocean | Render.com |
+| ------------------------------------------------ | --------------------------------------- | --------------------------------------- | ---------------------------------------------- | ---------------------------------------------------- |
| [![Deploy on Docker][docker-btn]][docker-deploy] | [![Deploy on AWS][aws-btn]][aws-deploy] | [![Deploy on GCP][gcp-btn]][gcp-deploy] | [![Deploy on DigitalOcean][do-btn]][do-deploy] | [![Deploy on Render.com][render-btn]][render-deploy] |
-| Railway | RepoCloud | Elestio |
-| --- | --- | --- |
+| Railway | RepoCloud | Elestio |
+| --------------------------------------------------- | --------------------------------------------------------- | --------------------------------------------------- |
| [![Deploy on Railway][railway-btn]][railway-deploy] | [![Deploy on RepoCloud][repocloud-btn]][repocloud-deploy] | [![Deploy on Elestio][elestio-btn]][elestio-deploy] |
@@ -224,7 +228,8 @@ AnythingLLM توسط Mintplex Labs Inc دارای ویژگی تلهمتری
میتوانید این ادعاها را با پیدا کردن تمام مکانهایی که `Telemetry.sendTelemetry` فراخوانی میشود، تأیید کنید. ارائهدهنده تلهمتری [PostHog](https://posthog.com/) است.
-[مشاهده همه رویدادهای تلهمتری در کد منبع](https://github.com/search?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry\(&type=code)
+[مشاهده همه رویدادهای تلهمتری در کد منبع](https://github.com/search?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry(&type=code)
+
@@ -259,10 +264,12 @@ AnythingLLM توسط Mintplex Labs Inc دارای ویژگی تلهمتری
---
+
Copyright © 2025 [Mintplex Labs][profile-link].
This project is [MIT](../LICENSE) licensed.
+
diff --git a/locales/README.ja-JP.md b/locales/README.ja-JP.md
index cd1798bd..dbd8c5d3 100644
--- a/locales/README.ja-JP.md
+++ b/locales/README.ja-JP.md
@@ -95,6 +95,8 @@ AnythingLLMは、ドキュメントを`ワークスペース`と呼ばれるオ
- [PPIO](https://ppinfra.com?utm_source=github_anything-llm)
- [CometAPI (チャットモデル)](https://api.cometapi.com/)
- [Docker Model Runner](https://docs.docker.com/ai/model-runner/)
+- [PrivateModeAI (chat models)](https://privatemode.ai/)
+- [SambaNova Cloud (chat models)](https://cloud.sambanova.ai/)
**埋め込みモデル：**
@@ -150,8 +152,8 @@ Mintplex Labsおよびコミュニティは、AnythingLLMをローカルで実
|----------------------------------------|----|-----|---------------|------------|
| [![Docker上でデプロイ][docker-btn]][docker-deploy] | [![AWS上でデプロイ][aws-btn]][aws-deploy] | [![GCP上でデプロイ][gcp-btn]][gcp-deploy] | [![DigitalOcean上でデプロイ][do-btn]][do-deploy] | [![Render.com上でデプロイ][render-btn]][render-deploy] |
-| Railway |
-| --------------------------------------------------- |
+| Railway |
+| ----------------------------------------------------- |
| [![Railway上でデプロイ][railway-btn]][railway-deploy] |
[Dockerを使用せずに本番環境のAnythingLLMインスタンスを設定する →](../BARE_METAL.md)
@@ -201,7 +203,7 @@ Mintplex Labs Inc.によって開発されたAnythingLLMには、匿名の使用
これらの主張を検証するには、`Telemetry.sendTelemetry`が呼び出されるすべての場所を見つけてください。また、これらのイベントは出力ログに書き込まれるため、送信された具体的なデータも確認できます。IPアドレスやその他の識別情報は収集されません。テレメトリープロバイダーは[PostHog](https://posthog.com/)です。
-[ソースコード内のすべてのテレメトリーイベントを表示](https://github.com/search?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry\(&type=code)
+[ソースコード内のすべてのテレメトリーイベントを表示](https://github.com/search?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry(&type=code)
diff --git a/locales/README.tr-TR.md b/locales/README.tr-TR.md
index 37f30018..e6c4cb3d 100644
--- a/locales/README.tr-TR.md
+++ b/locales/README.tr-TR.md
@@ -32,7 +32,6 @@ Belgelerinizle sohbet edin, yapay zeka ajanlarını kullanın, son derece özell
English ·
简体中文 ·
日本語 ·
Turkish
-
👉 Masaüstü için AnythingLLM (Mac, Windows ve Linux)! Şimdi İndir
@@ -87,7 +86,7 @@ AnythingLLM, belgelerinizi **"çalışma alanları" (workspaces)** adı verilen
- [LM Studio (all models)](https://lmstudio.ai)
- [LocalAi (all models)](https://localai.io/)
- [Together AI (chat models)](https://www.together.ai/)
-- [Fireworks AI (chat models)](https://fireworks.ai/)
+- [Fireworks AI (chat models)](https://fireworks.ai/)
- [Perplexity (chat models)](https://www.perplexity.ai/)
- [OpenRouter (chat models)](https://openrouter.ai/)
- [DeepSeek (chat models)](https://deepseek.com/)
@@ -103,6 +102,8 @@ AnythingLLM, belgelerinizi **"çalışma alanları" (workspaces)** adı verilen
- [Novita AI (chat models)](https://novita.ai/model-api/product/llm-api?utm_source=github_anything-llm&utm_medium=github_readme&utm_campaign=link)
- [PPIO](https://ppinfra.com?utm_source=github_anything-llm)
- [Docker Model Runner](https://docs.docker.com/ai/model-runner/)
+- [PrivateModeAI (chat models)](https://privatemode.ai/)
+- [SambaNova Cloud (chat models)](https://cloud.sambanova.ai/)
**Embedder modelleri:**
@@ -161,8 +162,8 @@ Mintplex Labs ve topluluk, AnythingLLM'i yerel olarak çalıştırmak için çe
|----------------------------------------|----|-----|---------------|------------|
| [![Deploy on Docker][docker-btn]][docker-deploy] | [![Deploy on AWS][aws-btn]][aws-deploy] | [![Deploy on GCP][gcp-btn]][gcp-deploy] | [![Deploy on DigitalOcean][do-btn]][do-deploy] | [![Deploy on Render.com][render-btn]][render-deploy] |
-| Railway | RepoCloud | Elestio |
-| --- | --- | --- |
+| Railway | RepoCloud | Elestio |
+| --------------------------------------------------- | --------------------------------------------------------- | --------------------------------------------------- |
| [![Deploy on Railway][railway-btn]][railway-deploy] | [![Deploy on RepoCloud][repocloud-btn]][repocloud-deploy] | [![Deploy on Elestio][elestio-btn]][elestio-deploy] |
[veya Docker kullanmadan üretim ortamında AnythingLLM kurun →](../BARE_METAL.md)
@@ -214,11 +215,10 @@ Yalnızca ürün ve yol haritası kararlarını almamıza yardımcı olacak kull
Bu verileri doğrulamak için kod içinde **`Telemetry.sendTelemetry` çağrılarını** inceleyebilirsiniz. Ayrıca, bu olaylar günlük kaydına yazıldığı için hangi verilerin gönderildiğini görebilirsiniz (eğer etkinleştirilmişse). **IP adresi veya diğer tanımlayıcı bilgiler toplanmaz.** Telemetri sağlayıcısı, açık kaynaklı bir telemetri toplama hizmeti olan [PostHog](https://posthog.com/)‘dur.
-[Kaynak kodda tüm telemetri olaylarını görüntüle](https://github.com/search?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry\(&type=code)
+[Kaynak kodda tüm telemetri olaylarını görüntüle](https://github.com/search?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry(&type=code)
-
## 👋 Katkıda Bulunma
- Bir **issue** oluşturun.
diff --git a/locales/README.zh-CN.md b/locales/README.zh-CN.md
index 4c978a9a..68add19a 100644
--- a/locales/README.zh-CN.md
+++ b/locales/README.zh-CN.md
@@ -54,6 +54,7 @@ AnythingLLM是一个全栈应用程序，您可以使用现成的商业大语言
AnythingLLM将您的文档划分为称为`workspaces` (工作区)的对象。工作区的功能类似于线程，同时增加了文档的容器化。工作区可以共享文档，但工作区之间的内容不会互相干扰或污染，因此您可以保持每个工作区的上下文清晰。
## AnythingLLM的一些酷炫特性
+
- 🆕 [**完全兼容 MCP**](https://docs.anythingllm.com/mcp-compatibility/overview)
- 🆕 [**无代码AI代理构建器**](https://docs.anythingllm.com/agent-flows/overview)
- 🖼️ **多用户实例支持和权限管理（支持封闭源和开源LLM！）**
@@ -103,6 +104,8 @@ AnythingLLM将您的文档划分为称为`workspaces` (工作区)的对象。工
- [PPIO (聊天模型)](https://ppinfra.com?utm_source=github_anything-llm)
- [CometAPI (聊天模型)](https://api.cometapi.com/)
- [Docker Model Runner](https://docs.docker.com/ai/model-runner/)
+- [PrivateModeAI (chat models)](https://privatemode.ai/)
+- [SambaNova Cloud (chat models)](https://cloud.sambanova.ai/)
**支持的嵌入模型：**
@@ -161,8 +164,8 @@ Mintplex Labs和社区维护了许多部署方法、脚本和模板，您可以
|----------------------------------------|----|-----|---------------|------------|
| [![在 Docker 上部署][docker-btn]][docker-deploy] | [![在 AWS 上部署][aws-btn]][aws-deploy] | [![在 GCP 上部署][gcp-btn]][gcp-deploy] | [![在DigitalOcean上部署][do-btn]][do-deploy] | [![在 Render.com 上部署][render-btn]][render-deploy] |
-| Railway | RepoCloud | Elestio |
-| --- | --- | --- |
+| Railway | RepoCloud | Elestio |
+| --------------------------------------------------- | --------------------------------------------------------- | --------------------------------------------------- |
| [![在 Railway 上部署][railway-btn]][railway-deploy] | [![在 RepoCloud 上部署][repocloud-btn]][repocloud-deploy] | [![在 Elestio 上部署][elestio-btn]][elestio-deploy] |
[其他方案：不使用Docker配置AnythingLLM实例 →](../BARE_METAL.md)
@@ -194,9 +197,6 @@ _以下是一些与 AnythingLLM 兼容的应用程序，但并非由 Mintplex La
有关 AnythingLLM 的远程信息收集与隐私保护更多信息
-
-
-
### 为什么收集信息？
我们使用这些信息来帮助我们理解 AnythingLLM 的使用情况，帮助我们确定新功能和错误修复的优先级，并帮助我们提高 AnythingLLM 的性能和稳定性。
@@ -219,7 +219,7 @@ _以下是一些与 AnythingLLM 兼容的应用程序，但并非由 Mintplex La
我们非常重视隐私，且不用烦人的弹窗问卷来获取反馈，希望你能理解为什么我们想要知道该工具的使用情况，这样我们才能打造真正值得使用的产品。所有匿名数据 _绝不会_ 与任何第三方共享。
-[在源代码中查看所有信息收集活动](https://github.com/search?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry\(&type=code)
+[在源代码中查看所有信息收集活动](https://github.com/search?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry(&type=code)
diff --git a/server/.env.example b/server/.env.example
index 6523d100..a1957878 100644
--- a/server/.env.example
+++ b/server/.env.example
@@ -170,6 +170,10 @@ SIG_SALT='salt' # Please generate random string at least 32 chars long.
# PRIVATEMODE_LLM_BASE_PATH='http://127.0.0.1:8080'
# PRIVATEMODE_LLM_MODEL_PREF='gemma-3-27b'
+# LLM_PROVIDER='sambanova'
+# SAMBANOVA_LLM_API_KEY='xxx-xxx-xxx'
+# SAMBANOVA_LLM_MODEL_PREF='gpt-oss-120b'
+
###########################################
######## Embedding API SElECTION ##########
###########################################
diff --git a/server/endpoints/utils.js b/server/endpoints/utils.js
index 61a61d6a..9580b043 100644
--- a/server/endpoints/utils.js
+++ b/server/endpoints/utils.js
@@ -165,6 +165,9 @@ function getModelTag() {
case "privatemode":
model = process.env.PRIVATEMODE_LLM_MODEL_PREF;
break;
+ case "sambanova":
+ model = process.env.SAMBANOVA_LLM_MODEL_PREF;
+ break;
default:
model = "--";
break;
diff --git a/server/models/systemSettings.js b/server/models/systemSettings.js
index 9b6e02ae..debb4b60 100644
--- a/server/models/systemSettings.js
+++ b/server/models/systemSettings.js
@@ -681,6 +681,10 @@ const SystemSettings = {
// Privatemode Keys
PrivateModeBasePath: process.env.PRIVATEMODE_LLM_BASE_PATH,
PrivateModeModelPref: process.env.PRIVATEMODE_LLM_MODEL_PREF,
+
+ // SambaNova Keys
+ SambaNovaLLMApiKey: !!process.env.SAMBANOVA_LLM_API_KEY,
+ SambaNovaLLMModelPref: process.env.SAMBANOVA_LLM_MODEL_PREF,
};
},
diff --git a/server/utils/AiProviders/modelMap/index.js b/server/utils/AiProviders/modelMap/index.js
index 5e191379..30ac871e 100644
--- a/server/utils/AiProviders/modelMap/index.js
+++ b/server/utils/AiProviders/modelMap/index.js
@@ -20,6 +20,7 @@ class ContextWindowFinder {
deepseek: "deepseek",
moonshot: "moonshot",
zai: "vercel_ai_gateway", // Vercel has correct context windows for Z.AI models
+ sambanova: "sambanova",
};
static expiryMs = 1000 * 60 * 60 * 24 * 3; // 3 days
static remoteUrl =
diff --git a/server/utils/AiProviders/sambanova/index.js b/server/utils/AiProviders/sambanova/index.js
new file mode 100644
index 00000000..49ae1614
--- /dev/null
+++ b/server/utils/AiProviders/sambanova/index.js
@@ -0,0 +1,268 @@
+const { NativeEmbedder } = require("../../EmbeddingEngines/native");
+const {
+ LLMPerformanceMonitor,
+} = require("../../helpers/chat/LLMPerformanceMonitor");
+const { v4: uuidv4 } = require("uuid");
+const { writeResponseChunk } = require("../../helpers/chat/responses");
+const { MODEL_MAP } = require("../modelMap");
+
+class SambaNovaLLM {
+ constructor(embedder = null, modelPreference = null) {
+ if (!process.env.SAMBANOVA_LLM_API_KEY)
+ throw new Error("No SambaNova API key was set.");
+ this.className = "SambaNovaLLM";
+ const { OpenAI: OpenAIApi } = require("openai");
+
+ this.openai = new OpenAIApi({
+ baseURL: "https://api.sambanova.ai/v1",
+ apiKey: process.env.SAMBANOVA_LLM_API_KEY,
+ });
+ this.model = modelPreference || process.env.SAMBANOVA_LLM_MODEL_PREF;
+ this.limits = {
+ history: this.promptWindowLimit() * 0.15,
+ system: this.promptWindowLimit() * 0.15,
+ user: this.promptWindowLimit() * 0.7,
+ };
+
+ this.embedder = embedder ?? new NativeEmbedder();
+ this.defaultTemp = 0.7;
+ this.log(
+ `Initialized ${this.model} with context window ${this.promptWindowLimit()}`
+ );
+ }
+
+ log(text, ...args) {
+ console.log(`\x1b[36m[${this.className}]\x1b[0m ${text}`, ...args);
+ }
+
+ #appendContext(contextTexts = []) {
+ if (!contextTexts || !contextTexts.length) return "";
+ return (
+ "\nContext:\n" +
+ contextTexts
+ .map((text, i) => {
+ return `[CONTEXT ${i}]:\n${text}\n[END CONTEXT ${i}]\n\n`;
+ })
+ .join("")
+ );
+ }
+
+ streamingEnabled() {
+ return "streamGetChatCompletion" in this;
+ }
+
+ static promptWindowLimit(modelName) {
+ return MODEL_MAP.get("sambanova", modelName) ?? 131072;
+ }
+
+ promptWindowLimit() {
+ return MODEL_MAP.get("sambanova", this.model) ?? 131072;
+ }
+
+ async isValidChatCompletionModel(modelName = "") {
+ return !!modelName; // name just needs to exist
+ }
+
+ /**
+ * Generates appropriate content array for a message + attachments.
+ * @param {{userPrompt:string, attachments: import("../../helpers").Attachment[]}}
+ * @returns {string|object[]}
+ */
+ #generateContent({ userPrompt, attachments = [] }) {
+ if (!attachments.length) return userPrompt;
+
+ const content = [{ type: "text", text: userPrompt }];
+ for (let attachment of attachments) {
+ content.push({
+ type: "image_url",
+ image_url: {
+ url: attachment.contentString,
+ },
+ });
+ }
+ return content.flat();
+ }
+
+ /**
+ * Construct the user prompt for this model.
+ * @param {{attachments: import("../../helpers").Attachment[]}} param0
+ * @returns
+ */
+ constructPrompt({
+ systemPrompt = "",
+ contextTexts = [],
+ chatHistory = [],
+ userPrompt = "",
+ attachments = [],
+ }) {
+ const prompt = {
+ role: "system",
+ content: `${systemPrompt}${this.#appendContext(contextTexts)}`,
+ };
+ return [
+ prompt,
+ ...chatHistory,
+ {
+ role: "user",
+ content: this.#generateContent({ userPrompt, attachments }),
+ },
+ ];
+ }
+
+ async getChatCompletion(messages = null, { temperature = 0.7 }) {
+ const result = await LLMPerformanceMonitor.measureAsyncFunction(
+ this.openai.chat.completions
+ .create({
+ model: this.model,
+ messages,
+ temperature,
+ })
+ .catch((e) => {
+ throw new Error(e.message);
+ })
+ );
+
+ if (
+ !result.output.hasOwnProperty("choices") ||
+ result.output.choices.length === 0
+ )
+ return null;
+
+ return {
+ textResponse: result.output.choices[0].message.content,
+ metrics: {
+ prompt_tokens: result.output.usage?.prompt_tokens || 0,
+ completion_tokens: result.output.usage?.completion_tokens || 0,
+ total_tokens: result.output.usage?.total_tokens || 0,
+ outputTps: result.output.usage?.total_tokens_per_sec || 0,
+ duration: result.duration,
+ model: this.model,
+ timestamp: new Date(),
+ },
+ };
+ }
+
+ async streamGetChatCompletion(messages = null, { temperature = 0.7 }) {
+ const measuredStreamRequest = await LLMPerformanceMonitor.measureStream({
+ func: this.openai.chat.completions.create({
+ model: this.model,
+ stream: true,
+ messages,
+ temperature,
+ stream_options: {
+ include_usage: true,
+ },
+ }),
+ messages,
+ runPromptTokenCalculation: false,
+ modelTag: this.model,
+ });
+
+ return measuredStreamRequest;
+ }
+
+ handleStream(response, stream, responseProps) {
+ const { uuid = uuidv4(), sources = [] } = responseProps;
+ let hasUsageMetrics = false;
+ let usage = {
+ prompt_tokens: 0,
+ total_tokens: 0,
+ outputTps: 0,
+ completion_tokens: 0,
+ };
+
+ return new Promise(async (resolve) => {
+ let fullText = "";
+ const handleAbort = () => {
+ stream?.endMeasurement(usage);
+ clientAbortedHandler(resolve, fullText);
+ };
+ response.on("close", handleAbort);
+
+ try {
+ for await (const chunk of stream) {
+ const message = chunk?.choices?.[0];
+ const token = message?.delta?.content;
+
+ if (
+ chunk.hasOwnProperty("usage") && // exists
+ !!chunk.usage &&
+ Object.values(chunk.usage).length > 0
+ ) {
+ if (chunk.usage.hasOwnProperty("prompt_tokens"))
+ usage.prompt_tokens = Number(chunk.usage.prompt_tokens);
+ if (chunk.usage.hasOwnProperty("completion_tokens"))
+ usage.completion_tokens = Number(chunk.usage.completion_tokens);
+ if (chunk.usage.hasOwnProperty("total_tokens"))
+ usage.total_tokens = Number(chunk.usage.total_tokens);
+ if (chunk.usage.hasOwnProperty("total_tokens_per_sec"))
+ usage.outputTps = Number(chunk.usage.total_tokens_per_sec);
+ hasUsageMetrics = true;
+ }
+
+ if (token) {
+ fullText += token;
+ if (!hasUsageMetrics) usage.completion_tokens++;
+ writeResponseChunk(response, {
+ uuid,
+ sources: [],
+ type: "textResponseChunk",
+ textResponse: token,
+ close: false,
+ error: false,
+ });
+ }
+
+ if (
+ message?.hasOwnProperty("finish_reason") &&
+ message.finish_reason !== "" &&
+ message.finish_reason !== null
+ ) {
+ writeResponseChunk(response, {
+ uuid,
+ sources,
+ type: "textResponseChunk",
+ textResponse: "",
+ close: true,
+ error: false,
+ });
+ }
+ }
+
+ response.removeListener("close", handleAbort);
+ stream?.endMeasurement(usage);
+ resolve(fullText);
+ } catch (e) {
+ this.log(`\x1b[43m\x1b[34m[STREAMING ERROR]\x1b[0m ${e.message}`);
+ writeResponseChunk(response, {
+ uuid,
+ type: "abort",
+ textResponse: null,
+ sources: [],
+ close: true,
+ error: e.message,
+ });
+ stream?.endMeasurement(usage);
+ resolve(fullText);
+ }
+ });
+ }
+
+ // Simple wrapper for dynamic embedder & normalize interface for all LLM implementations
+ async embedTextInput(textInput) {
+ return await this.embedder.embedTextInput(textInput);
+ }
+ async embedChunks(textChunks = []) {
+ return await this.embedder.embedChunks(textChunks);
+ }
+
+ async compressMessages(promptArgs = {}, rawHistory = []) {
+ const { messageArrayCompressor } = require("../../helpers/chat");
+ const messageArray = this.constructPrompt(promptArgs);
+ return await messageArrayCompressor(this, messageArray, rawHistory);
+ }
+}
+
+module.exports = {
+ SambaNovaLLM,
+};
diff --git a/server/utils/agents/aibitat/index.js b/server/utils/agents/aibitat/index.js
index 86e1a214..fbcdb866 100644
--- a/server/utils/agents/aibitat/index.js
+++ b/server/utils/agents/aibitat/index.js
@@ -996,6 +996,8 @@ ${this.getHistory({ to: route.to })
return new Providers.DockerModelRunnerProvider({ model: config.model });
case "privatemode":
return new Providers.PrivatemodeProvider({ model: config.model });
+ case "sambanova":
+ return new Providers.SambaNovaProvider({ model: config.model });
default:
throw new Error(
`Unknown provider: ${config.provider}. Please use a valid provider.`
diff --git a/server/utils/agents/aibitat/providers/ai-provider.js b/server/utils/agents/aibitat/providers/ai-provider.js
index d1da6bae..7b10e729 100644
--- a/server/utils/agents/aibitat/providers/ai-provider.js
+++ b/server/utils/agents/aibitat/providers/ai-provider.js
@@ -257,6 +257,14 @@ class Provider {
apiKey: null,
...config,
});
+ case "sambanova":
+ return new ChatOpenAI({
+ configuration: {
+ baseURL: "https://api.sambanova.ai/v1",
+ },
+ apiKey: process.env.SAMBANOVA_LLM_API_KEY ?? null,
+ ...config,
+ });
// OSS Model Runners
// case "anythingllm_ollama":
// return new ChatOllama({
diff --git a/server/utils/agents/aibitat/providers/index.js b/server/utils/agents/aibitat/providers/index.js
index 1862dc66..6c6fbfb0 100644
--- a/server/utils/agents/aibitat/providers/index.js
+++ b/server/utils/agents/aibitat/providers/index.js
@@ -31,6 +31,7 @@ const GiteeAIProvider = require("./giteeai.js");
const CohereProvider = require("./cohere.js");
const DockerModelRunnerProvider = require("./dockerModelRunner.js");
const PrivatemodeProvider = require("./privatemode.js");
+const SambaNovaProvider = require("./sambanova.js");
module.exports = {
OpenAIProvider,
@@ -66,4 +67,5 @@ module.exports = {
CohereProvider,
DockerModelRunnerProvider,
PrivatemodeProvider,
+ SambaNovaProvider,
};
diff --git a/server/utils/agents/aibitat/providers/sambanova.js b/server/utils/agents/aibitat/providers/sambanova.js
new file mode 100644
index 00000000..75b307d1
--- /dev/null
+++ b/server/utils/agents/aibitat/providers/sambanova.js
@@ -0,0 +1,85 @@
+const OpenAI = require("openai");
+const Provider = require("./ai-provider.js");
+const InheritMultiple = require("./helpers/classes.js");
+const UnTooled = require("./helpers/untooled.js");
+
+class SambaNovaProvider extends InheritMultiple([Provider, UnTooled]) {
+ model;
+
+ constructor(config = {}) {
+ const { model = "Meta-Llama-3.1-8B-Instruct" } = config;
+ super();
+ const client = new OpenAI({
+ baseURL: "https://api.sambanova.ai/v1",
+ apiKey: process.env.SAMBANOVA_LLM_API_KEY,
+ maxRetries: 3,
+ });
+
+ this._client = client;
+ this.model = model;
+ this.verbose = true;
+ }
+
+ /**
+ * Create a completion based on the received messages.
+ *
+ * @param messages A list of messages to send to the API.
+ * @param functions
+ * @returns The completion.
+ */
+ get client() {
+ return this._client;
+ }
+
+ get supportsAgentStreaming() {
+ return true;
+ }
+
+ async #handleFunctionCallChat({ messages = [] }) {
+ return await this.client.chat.completions
+ .create({
+ model: this.model,
+ messages,
+ })
+ .then((result) => {
+ if (!result.hasOwnProperty("choices"))
+ throw new Error("SambaNova chat: No results!");
+ if (result.choices.length === 0)
+ throw new Error("SambaNova chat: No results length!");
+ return result.choices[0].message.content;
+ });
+ }
+
+ async #handleFunctionCallStream({ messages = [] }) {
+ return await this.client.chat.completions.create({
+ model: this.model,
+ stream: true,
+ messages,
+ });
+ }
+
+ async stream(messages, functions = [], eventHandler = null) {
+ return await UnTooled.prototype.stream.call(
+ this,
+ messages,
+ functions,
+ this.#handleFunctionCallStream.bind(this),
+ eventHandler
+ );
+ }
+
+ async complete(messages, functions = []) {
+ return await UnTooled.prototype.complete.call(
+ this,
+ messages,
+ functions,
+ this.#handleFunctionCallChat.bind(this)
+ );
+ }
+
+ getCost(_usage) {
+ return 0;
+ }
+}
+
+module.exports = SambaNovaProvider;
diff --git a/server/utils/agents/index.js b/server/utils/agents/index.js
index 582d5495..8712b2cc 100644
--- a/server/utils/agents/index.js
+++ b/server/utils/agents/index.js
@@ -229,6 +229,10 @@ class AgentHandler {
"Privatemode base path must be provided to use agents."
);
break;
+ case "sambanova":
+ if (!process.env.SAMBANOVA_LLM_API_KEY)
+ throw new Error("SambaNova API key must be provided to use agents.");
+ break;
default:
throw new Error(
"No workspace agent provider set. Please set your agent provider in the workspace's settings"
@@ -313,6 +317,8 @@ class AgentHandler {
return process.env.DOCKER_MODEL_RUNNER_LLM_MODEL_PREF ?? null;
case "privatemode":
return process.env.PRIVATEMODE_LLM_MODEL_PREF ?? null;
+ case "sambanova":
+ return process.env.SAMBANOVA_LLM_MODEL_PREF ?? null;
default:
return null;
}
diff --git a/server/utils/helpers/customModels.js b/server/utils/helpers/customModels.js
index 974099af..17fe056e 100644
--- a/server/utils/helpers/customModels.js
+++ b/server/utils/helpers/customModels.js
@@ -46,6 +46,7 @@ const SUPPORT_CUSTOM_MODELS = [
"giteeai",
"docker-model-runner",
"privatemode",
+ "sambanova",
// Embedding Engines
"native-embedder",
"cohere-embedder",
@@ -123,6 +124,8 @@ async function getCustomModels(provider = "", apiKey = null, basePath = null) {
return await getDockerModelRunnerModels(basePath);
case "privatemode":
return await getPrivatemodeModels(basePath, "generate");
+ case "sambanova":
+ return await getSambaNovaModels(apiKey);
default:
return { models: [], error: "Invalid provider for custom models" };
}
@@ -932,6 +935,53 @@ async function getPrivatemodeModels(basePath = null, task = "any") {
}
}
+/**
+ * Get SambaNova models
+ * @param {string} _apiKey - The API key to use
+ * @returns {Promise<{models: Array<{id: string, organization: string, name: string}>, error: string | null}>}
+ */
+async function getSambaNovaModels(_apiKey = null) {
+ try {
+ const apiKey =
+ _apiKey === true
+ ? process.env.SAMBANOVA_LLM_API_KEY
+ : _apiKey || process.env.SAMBANOVA_LLM_API_KEY || null;
+ const { OpenAI: OpenAIApi } = require("openai");
+ const openai = new OpenAIApi({
+ baseURL: "https://api.sambanova.ai/v1",
+ apiKey,
+ });
+ const models = await openai.models
+ .list()
+ .then((results) => results.data)
+ .then((models) =>
+ models.filter((model) => !model.id.toLowerCase().startsWith("whisper"))
+ )
+ .then((models) =>
+ models.map((model) => {
+ const organization =
+ model.hasOwnProperty("owned_by") &&
+ model.owned_by !== "no-reply@sambanova.ai"
+ ? model.owned_by
+ : "SambaNova";
+ return {
+ id: model.id,
+ organization,
+ name: model.id,
+ };
+ })
+ )
+ .catch((e) => {
+ console.error(`SambaNova:listModels`, e.message);
+ return [];
+ });
+ return { models, error: null };
+ } catch (e) {
+ console.error(`SambaNova:getSambaNovaModels`, e.message);
+ return { models: [], error: "Could not fetch SambaNova Models" };
+ }
+}
+
module.exports = {
getCustomModels,
SUPPORT_CUSTOM_MODELS,
diff --git a/server/utils/helpers/index.js b/server/utils/helpers/index.js
index 38f82306..fba0feea 100644
--- a/server/utils/helpers/index.js
+++ b/server/utils/helpers/index.js
@@ -237,6 +237,9 @@ function getLLMProvider({ provider = null, model = null } = {}) {
case "privatemode":
const { PrivatemodeLLM } = require("../AiProviders/privatemode");
return new PrivatemodeLLM(embedder, model);
+ case "sambanova":
+ const { SambaNovaLLM } = require("../AiProviders/sambanova");
+ return new SambaNovaLLM(embedder, model);
default:
throw new Error(
`ENV: No valid LLM_PROVIDER value found in environment! Using ${process.env.LLM_PROVIDER}`
@@ -410,6 +413,9 @@ function getLLMProviderClass({ provider = null } = {}) {
case "privatemode":
const { PrivateModeLLM } = require("../AiProviders/privatemode");
return PrivateModeLLM;
+ case "sambanova":
+ const { SambaNovaLLM } = require("../AiProviders/sambanova");
+ return SambaNovaLLM;
default:
return null;
}
@@ -490,6 +496,8 @@ function getBaseLLMProviderModel({ provider = null } = {}) {
return process.env.DOCKER_MODEL_RUNNER_LLM_MODEL_PREF;
case "privatemode":
return process.env.PRIVATEMODE_LLM_MODEL_PREF;
+ case "sambanova":
+ return process.env.SAMBANOVA_LLM_MODEL_PREF;
default:
return null;
}
diff --git a/server/utils/helpers/updateENV.js b/server/utils/helpers/updateENV.js
index fdac6018..c6ed6c52 100644
--- a/server/utils/helpers/updateENV.js
+++ b/server/utils/helpers/updateENV.js
@@ -809,6 +809,16 @@ const KEY_MAPPING = {
envKey: "PRIVATEMODE_LLM_MODEL_PREF",
checks: [isNotEmpty],
},
+
+ // SambaNova Options
+ SambaNovaLLMApiKey: {
+ envKey: "SAMBANOVA_LLM_API_KEY",
+ checks: [isNotEmpty],
+ },
+ SambaNovaLLMModelPref: {
+ envKey: "SAMBANOVA_LLM_MODEL_PREF",
+ checks: [isNotEmpty],
+ },
};
function isNotEmpty(input = "") {
@@ -924,6 +934,7 @@ function supportedLLM(input = "") {
"giteeai",
"docker-model-runner",
"privatemode",
+ "sambanova",
].includes(input);
return validSelection ? null : `${input} is not a valid LLM provider.`;
}