Merge pull request #591 from mindcraft-bots/cerebras

Cerebras and better default models

README.md

@@ -12,7 +12,8 @@ Do not connect this bot to public servers with coding enabled. This project allo
 
 - [Minecraft Java Edition](https://www.minecraft.net/en-us/store/minecraft-java-bedrock-edition-pc) (up to v1.21.1, recommend v1.21.1)
 - [Node.js Installed](https://nodejs.org/) (at least v18)
-- One of these: [OpenAI API Key](https://openai.com/blog/openai-api) | [Gemini API Key](https://aistudio.google.com/app/apikey) | [Anthropic API Key](https://docs.anthropic.com/claude/docs/getting-access-to-claude) | [Replicate API Key](https://replicate.com/) | [Hugging Face API Key](https://huggingface.co/) | [Groq API Key](https://console.groq.com/keys) | [Ollama Installed](https://ollama.com/download). | [Mistral API Key](https://docs.mistral.ai/getting-started/models/models_overview/) | [Qwen API Key [Intl.]](https://www.alibabacloud.com/help/en/model-studio/developer-reference/get-api-key)/[[cn]](https://help.aliyun.com/zh/model-studio/getting-started/first-api-call-to-qwen?) | [Novita AI API Key](https://novita.ai/settings?utm_source=github_mindcraft&utm_medium=github_readme&utm_campaign=link#key-management) |
+- One of these: [OpenAI API Key](https://openai.com/blog/openai-api) | [Gemini API Key](https://aistudio.google.com/app/apikey) | [Anthropic API Key](https://docs.anthropic.com/claude/docs/getting-access-to-claude) | [Replicate API Key](https://replicate.com/) | [Hugging Face API Key](https://huggingface.co/) | [Groq API Key](https://console.groq.com/keys) | [Ollama Installed](https://ollama.com/download). | [Mistral API Key](https://docs.mistral.ai/getting-started/models/models_overview/) | [Qwen API Key [Intl.]](https://www.alibabacloud.com/help/en/model-studio/developer-reference/get-api-key)/[[cn]](https://help.aliyun.com/zh/model-studio/getting-started/first-api-call-to-qwen?) | [Novita AI API Key](https://novita.ai/settings?utm_source=github_mindcraft&utm_medium=github_readme&utm_campaign=link#key-management) | [Cerebras API Key](https://cloud.cerebras.ai)
+
 
 ## Install and Run
 
@@ -64,6 +65,7 @@ You can configure the agent's name, model, and prompts in their profile like `an
 | `glhf.chat` | `GHLF_API_KEY` | `glhf/hf:meta-llama/Llama-3.1-405B-Instruct` | [docs](https://glhf.chat/user-settings/api) |
 | `hyperbolic` | `HYPERBOLIC_API_KEY` | `hyperbolic/deepseek-ai/DeepSeek-V3` | [docs](https://docs.hyperbolic.xyz/docs/getting-started) |
 | `vllm` | n/a | `vllm/llama3` | n/a |
+| `cerebras` | `CEREBRAS_API_KEY` | `cerebras/llama-3.3-70b` | [docs](https://inference-docs.cerebras.ai/introduction) |
 
 If you use Ollama, to install the models used by default (generation and embedding), execute the following terminal command:
 `ollama pull llama3.1 && ollama pull nomic-embed-text`

keys.example.json

@@ -13,5 +13,6 @@
     "GHLF_API_KEY": "",
     "HYPERBOLIC_API_KEY": "",
     "NOVITA_API_KEY": "",
-    "OPENROUTER_API_KEY": ""
+    "OPENROUTER_API_KEY": "",
+    "CEREBRAS_API_KEY": ""
 }

package.json

@@ -2,6 +2,7 @@
     "type": "module",
     "dependencies": {
         "@anthropic-ai/sdk": "^0.17.1",
+        "@cerebras/cerebras_cloud_sdk": "^1.46.0",
         "@google/generative-ai": "^0.2.1",
         "@huggingface/inference": "^2.8.1",
         "@mistralai/mistralai": "^1.1.0",

profiles/claude.json

@@ -1,7 +1,7 @@
 {
     "name": "claude",
 
-    "model": "claude-3-5-sonnet-latest",
+    "model": "claude-4-sonnet-latest",
     
     "embedding": "openai"
 }

profiles/claude_thinker.json

@@ -2,7 +2,7 @@
     "name": "claude_thinker",
 
     "model": {
-        "model": "claude-3-7-sonnet-latest",
+        "model": "claude-4-sonnet-latest",
         "params": {
             "thinking": {
                 "type": "enabled",

profiles/gemini.json

@@ -1,7 +1,7 @@
 {
     "name": "gemini",
 
-    "model": "gemini-2.0-flash",
+    "model": "gemini-2.5-flash",
 
     "cooldown": 5000
 }

profiles/grok.json

@@ -1,7 +1,7 @@
 {
     "name": "Grok",
 
-    "model": "grok-beta",
+    "model": "grok-3-mini-latest",
     
     "embedding": "openai"
 }

src/models/cerebras.js

@@ -0,0 +1,61 @@
+import CerebrasSDK from '@cerebras/cerebras_cloud_sdk';
+import { strictFormat } from '../utils/text.js';
+import { getKey } from '../utils/keys.js';
+
+export class Cerebras {
+    static prefix = 'cerebras';
+    constructor(model_name, url, params) {
+        this.model_name = model_name;
+        this.url = url;
+        this.params = params;
+
+        // Initialize client with API key
+        this.client = new CerebrasSDK({ apiKey: getKey('CEREBRAS_API_KEY') });
+    }
+
+    async sendRequest(turns, systemMessage, stop_seq = '***') {
+        // Format messages array
+        const messages = strictFormat(turns);
+        messages.unshift({ role: 'system', content: systemMessage });
+
+        const pack = {
+            model: this.model_name || 'gpt-oss-120b',
+            messages,
+            stream: false,
+            ...(this.params || {}),
+        };
+
+        let res;
+        try {
+            const completion = await this.client.chat.completions.create(pack);
+            // OpenAI-compatible shape
+            res = completion.choices?.[0]?.message?.content || '';
+        } catch (err) {
+            console.error('Cerebras API error:', err);
+            res = 'My brain disconnected, try again.';
+        }
+        return res;
+    }
+
+    async sendVisionRequest(messages, systemMessage, imageBuffer) {
+        const imageMessages = [...messages];
+        imageMessages.push({
+            role: "user",
+            content: [
+                { type: "text", text: systemMessage },
+                {
+                    type: "image_url",
+                    image_url: {
+                        url: `data:image/jpeg;base64,${imageBuffer.toString('base64')}`
+                    }
+                }
+            ]
+        });
+        
+        return this.sendRequest(imageMessages, systemMessage);
+    }
+    
+    async embed(text) {
+        throw new Error('Embeddings are not supported by Cerebras.');
+    }
+}

src/models/gemini.js

@@ -37,7 +37,7 @@ export class Gemini {
     async sendRequest(turns, systemMessage) {
         let model;
         const modelConfig = {
-            model: this.model_name || "gemini-1.5-flash",
+            model: this.model_name || "gemini-2.5-flash",
             // systemInstruction does not work bc google is trash
         };
         if (this.url) {

src/models/grok.js

@@ -3,7 +3,7 @@ import { getKey } from '../utils/keys.js';
 
 // xAI doesn't supply a SDK for their models, but fully supports OpenAI and Anthropic SDKs
 export class Grok {
-    static prefix = 'grok';
+    static prefix = 'xai';
     constructor(model_name, url, params) {
         this.model_name = model_name;
         this.url = url;
@@ -20,13 +20,12 @@ export class Grok {
         this.openai = new OpenAIApi(config);
     }
 
-    async sendRequest(turns, systemMessage, stop_seq='***') {
+    async sendRequest(turns, systemMessage) {
         let messages = [{'role': 'system', 'content': systemMessage}].concat(turns);
 
         const pack = {
-            model: this.model_name || "grok-beta",
+            model: this.model_name || "grok-3-mini-latest",
             messages,
-            stop: [stop_seq],
             ...(this.params || {})
         };
 
@@ -43,7 +42,7 @@ export class Grok {
         catch (err) {
             if ((err.message == 'Context length exceeded' || err.code == 'context_length_exceeded') && turns.length > 1) {
                 console.log('Context length exceeded, trying again with shorter context.');
-                return await this.sendRequest(turns.slice(1), systemMessage, stop_seq);
+                return await this.sendRequest(turns.slice(1), systemMessage);
             } else if (err.message.includes('The model expects a single `text` element per message.')) {
                 console.log(err);
                 res = 'Vision is only supported by certain models.';

src/models/groq.js

@@ -50,7 +50,7 @@ export class GroqCloudAPI {
 
             let completion = await this.groq.chat.completions.create({
                 "messages": messages,
-                "model": this.model_name || "llama-3.3-70b-versatile",
+                "model": this.model_name || "qwen/qwen3-32b",
                 "stream": false,
                 "stop": stop_seq,
                 ...(this.params || {})

src/models/novita.js

@@ -26,7 +26,7 @@ export class Novita {
       messages = strictFormat(messages);
       
       const pack = {
-          model: this.model_name || "meta-llama/llama-3.1-70b-instruct",
+          model: this.model_name || "meta-llama/llama-4-scout-17b-16e-instruct",
           messages,
           stop: [stop_seq],
           ...(this.params || {})