@qvac/sdk 0.14.1 → 0.16.0

package/dist/examples/logging-streaming.js

@@ -1,10 +1,10 @@
-import { loadModel, completion, unloadModel, loggingStream, SDK_LOG_ID, LLAMA_3_2_1B_INST_Q4_0, GTE_LARGE_FP16, VERBOSITY, embed, } from "@qvac/sdk";
+import { loadModel, completion, unloadModel, loggingStream, SDK_LOG_ID, LLAMA_3_2_1B_INST_Q4_0, GTE_LARGE_FP16, VERBOSITY, embed } from '@qvac/sdk';
 try {
-    console.log("▸ Streaming SDK and model logs");
+    console.log('▸ Streaming SDK and model logs');
     // Note: To configure logging (level and console output), use config file:
     // { "loggerLevel": "debug", "loggerConsoleOutput": false } in qvac.config.json/js/ts
     // Subscribe to SDK server logs in background
-    console.log("▸ Subscribing to SDK server logs");
+    console.log('▸ Subscribing to SDK server logs');
     (async () => {
         for await (const log of loggingStream({ id: SDK_LOG_ID })) {
             console.log(`[SDK] [${log.level.toUpperCase()}] [${log.namespace}] ${log.message}`);
@@ -13,19 +13,19 @@ try {
         // Stream terminated - normal on shutdown
     });
     // Load models
-    console.log("▸ Loading models");
+    console.log('▸ Loading models');
     const llmModelId = await loadModel({
         modelSrc: LLAMA_3_2_1B_INST_Q4_0,
         modelConfig: {
             ctx_size: 2048,
             temp: 0.7,
-            verbosity: VERBOSITY.ERROR, // Only log errors, remaining logs are captured by loggingStream
-        },
+            verbosity: VERBOSITY.ERROR // Only log errors, remaining logs are captured by loggingStream
+        }
     });
     const embedModelId = await loadModel({
-        modelSrc: GTE_LARGE_FP16,
+        modelSrc: GTE_LARGE_FP16
     });
-    console.log("▸ Subscribing to model logs");
+    console.log('▸ Subscribing to model logs');
     (async () => {
         for await (const log of loggingStream({ id: llmModelId })) {
             const timestamp = new Date(log.timestamp).toISOString();
@@ -42,29 +42,31 @@ try {
     })().catch(() => {
         // Stream terminated - this is normal when model unloads
     });
-    const messages = [
-        { role: "user", content: "Count from 1 to 5 and explain each number." },
-    ];
+    const messages = [{ role: 'user', content: 'Count from 1 to 5 and explain each number.' }];
     const result = completion({
         modelId: llmModelId,
         history: messages,
         stream: true,
+        // Bound the generation: at temp 0.7 a 1B model can ignore "to 5" and ramble
+        // past ctx_size (2048) into a context overflow. This demo only needs enough
+        // tokens to show logs streaming.
+        generationParams: { predict: 512 }
     });
     const { embedding } = await embed({
         modelId: embedModelId,
-        text: messages[0]?.content ?? "Hello, world!",
+        text: messages[0]?.content ?? 'Hello, world!'
     });
-    console.log("▸ Response");
+    console.log('▸ Response');
     for await (const token of result.tokenStream) {
         process.stdout.write(token);
     }
-    console.log("Embedding (first 20 elements)", embedding.slice(0, 20));
-    console.log("Embeddings length", embedding.length);
-    console.log("▸ Three streams running: [SDK] server, [LLM] inference, [EMBED] embedding");
+    console.log('Embedding (first 20 elements)', embedding.slice(0, 20));
+    console.log('Embeddings length', embedding.length);
+    console.log('▸ Three streams running: [SDK] server, [LLM] inference, [EMBED] embedding');
     await unloadModel({ modelId: llmModelId, clearStorage: false });
     await unloadModel({ modelId: embedModelId, clearStorage: false });
 }
 catch (error) {
-    console.error("✖", error);
+    console.error('✖', error);
     process.exit(1);
 }

package/dist/examples/mcp-websearch.js

@@ -9,18 +9,18 @@
  *
  * Run with: bun run examples/mcp-websearch.ts
  */
-import { completion, loadModel, unloadModel, QWEN3_1_7B_INST_Q4, } from "../index.js";
+import { completion, loadModel, unloadModel, QWEN3_1_7B_INST_Q4 } from '../index.js';
 // MCP SDK is a user-installed optional dependency
 // Install with: bun add @modelcontextprotocol/sdk
 // eslint-disable-next-line import/no-unresolved
-import { Client } from "@modelcontextprotocol/sdk/client/index.js";
+import { Client } from '@modelcontextprotocol/sdk/client/index.js';
 // eslint-disable-next-line import/no-unresolved
-import { StdioClientTransport } from "@modelcontextprotocol/sdk/client/stdio.js";
+import { StdioClientTransport } from '@modelcontextprotocol/sdk/client/stdio.js';
 function parseSearchResults(mcpResult) {
     try {
         const result = mcpResult;
         // Extract text content from MCP response
-        const textContent = result.content?.find((c) => c.type === "text");
+        const textContent = result.content?.find((c) => c.type === 'text');
         if (!textContent?.text) {
             return JSON.stringify(mcpResult);
         }
@@ -28,55 +28,55 @@ function parseSearchResults(mcpResult) {
         const rawResults = JSON.parse(textContent.text);
         // Extract just the useful fields (title, url, snippet)
         const cleanResults = rawResults.slice(0, 5).map((r) => ({
-            title: r.title ?? "Unknown",
-            url: r.url ?? "",
-            snippet: r.snippet ?? "",
+            title: r.title ?? 'Unknown',
+            url: r.url ?? '',
+            snippet: r.snippet ?? ''
         }));
         // Format as concise text for LLM
         return cleanResults
             .map((r, i) => `[${i + 1}] ${r.title}\n    URL: ${r.url}\n    ${r.snippet}`)
-            .join("\n\n");
+            .join('\n\n');
     }
     catch {
         // If parsing fails, return a truncated version
-        const str = typeof mcpResult === "string" ? mcpResult : JSON.stringify(mcpResult);
+        const str = typeof mcpResult === 'string' ? mcpResult : JSON.stringify(mcpResult);
         return str.slice(0, 2000);
     }
 }
 let mcpClient = null;
 try {
-    console.log("▸ MCP DuckDuckGo Search Example\n");
+    console.log('▸ MCP DuckDuckGo Search Example\n');
     // ============================================================
     // STEP 1: Connect to DuckDuckGo MCP server
     // ============================================================
-    console.log("▸ Starting DuckDuckGo MCP server...");
+    console.log('▸ Starting DuckDuckGo MCP server...');
     mcpClient = new Client({
-        name: "qvac-ddg-example",
-        version: "1.0.0",
+        name: 'qvac-ddg-example',
+        version: '1.0.0'
     });
     const transport = new StdioClientTransport({
-        command: "npx",
-        args: ["-y", "@oevortex/ddg_search"],
+        command: 'npx',
+        args: ['-y', '@oevortex/ddg_search']
     });
     await mcpClient.connect(transport);
-    console.log("▸ MCP server connected\n");
+    console.log('▸ MCP server connected\n');
     // ============================================================
     // STEP 2: Load model
     // ============================================================
-    console.log("▸ Loading model...");
+    console.log('▸ Loading model...');
     const modelId = await loadModel({
         modelSrc: QWEN3_1_7B_INST_Q4,
         modelConfig: {
             ctx_size: 4096,
-            tools: true,
+            tools: true
         },
         onProgress: (p) => {
             const mb = (n) => (n / 1e6).toFixed(1);
             const line = `▸ Downloading ${p.percentage.toFixed(0)}% (${mb(p.downloaded)}/${mb(p.total)} MB)`;
             process.stderr.write(process.stderr.isTTY ? `\r${line}` : `${line}\n`);
             if (p.percentage >= 100)
-                process.stderr.write("\n");
-        },
+                process.stderr.write('\n');
+        }
     });
     console.log(`\n▸ Model loaded\n`);
     // ============================================================
@@ -84,35 +84,35 @@ try {
     // ============================================================
     const history = [
         {
-            role: "system",
+            role: 'system',
             content: `You are a helpful assistant with access to web search.
 Use the search tool when you need current information.
-Always cite your sources with the URL.`,
+Always cite your sources with the URL.`
         },
         {
-            role: "user",
-            content: "What is the current weather in New York City?",
-        },
+            role: 'user',
+            content: 'What is the current weather in New York City?'
+        }
     ];
-    console.log("▸ Asking AI to search the web...\n");
-    console.log("▸ AI Response:");
+    console.log('▸ Asking AI to search the web...\n');
+    console.log('▸ AI Response:');
     // Pass MCP client directly to completion - tools are adapted internally!
     const result = completion({
         modelId,
         history,
         stream: true,
-        mcp: [{ client: mcpClient, includeResources: false }],
+        mcp: [{ client: mcpClient, includeResources: false }]
     });
     for await (const token of result.tokenStream) {
         process.stdout.write(token);
     }
     const toolCalls = await result.toolCalls;
-    console.log("\n");
+    console.log('\n');
     // ============================================================
     // STEP 4: Execute tool calls using call() - automatic MCP routing!
     // ============================================================
     if (toolCalls.length > 0) {
-        console.log("▸ Executing search...\n");
+        console.log('▸ Executing search...\n');
         const toolResults = [];
         for (const toolCall of toolCalls) {
             console.log(`▸ ${toolCall.name}(${JSON.stringify(toolCall.arguments)})`);
@@ -126,49 +126,49 @@ Always cite your sources with the URL.`,
             const cleanResult = parseSearchResults(mcpResult);
             console.log(`▸ Got search results:`);
             console.log(cleanResult
-                .split("\n")
+                .split('\n')
                 .map((l) => `      ${l}`)
-                .join("\n"));
+                .join('\n'));
             console.log();
             toolResults.push({ id: toolCall.id, result: cleanResult });
         }
         // ============================================================
         // STEP 5: Continue with search results
         // ============================================================
-        console.log("▸ Getting AI response with search results...\n");
+        console.log('▸ Getting AI response with search results...\n');
         history.push({
-            role: "assistant",
-            content: await result.text,
+            role: 'assistant',
+            content: await result.text
         });
         for (const tr of toolResults) {
             history.push({
-                role: "tool",
-                content: tr.result,
+                role: 'tool',
+                content: tr.result
             });
         }
-        console.log("▸ Final Response:");
+        console.log('▸ Final Response:');
         const finalResult = completion({
             modelId,
             history,
             stream: true,
-            mcp: [{ client: mcpClient, includeResources: false }],
+            mcp: [{ client: mcpClient, includeResources: false }]
         });
         for await (const token of finalResult.tokenStream) {
             process.stdout.write(token);
         }
-        console.log("\n");
+        console.log('\n');
     }
     // ============================================================
     // Cleanup
     // ============================================================
-    console.log("▸ Cleaning up...");
+    console.log('▸ Cleaning up...');
     await unloadModel({ modelId, clearStorage: false });
-    console.log("▸ Done\n");
-    console.log("▸ Example completed!");
+    console.log('▸ Done\n');
+    console.log('▸ Example completed!');
     process.exit(0);
 }
 catch (error) {
-    console.error("✖", error);
+    console.error('✖', error);
     process.exit(1);
 }
 finally {

package/dist/examples/multi-model-demo.js

@@ -1,78 +1,78 @@
-import { completion, LLAMA_3_2_1B_INST_Q4_0, loadModel, QWEN3_1_7B_INST_Q4, unloadModel, } from "@qvac/sdk";
+import { completion, LLAMA_3_2_1B_INST_Q4_0, loadModel, QWEN3_1_7B_INST_Q4, unloadModel } from '@qvac/sdk';
 try {
     const alice = await loadModel({
         modelSrc: QWEN3_1_7B_INST_Q4,
         modelConfig: {
-            ctx_size: 4096,
-        },
+            ctx_size: 4096
+        }
     });
     const bob = await loadModel({
         modelSrc: LLAMA_3_2_1B_INST_Q4_0,
         modelConfig: {
-            ctx_size: 4096,
+            ctx_size: 4096
         },
         onProgress: (p) => {
             const mb = (n) => (n / 1e6).toFixed(1);
             const line = `▸ Downloading ${p.percentage.toFixed(0)}% (${mb(p.downloaded)}/${mb(p.total)} MB)`;
             process.stderr.write(process.stderr.isTTY ? `\r${line}` : `${line}\n`);
             if (p.percentage >= 100)
-                process.stderr.write("\n");
-        },
+                process.stderr.write('\n');
+        }
     });
     const aliceHistory = [
         {
-            role: "system",
-            content: "You answer questions about.",
+            role: 'system',
+            content: 'You answer questions about.'
         },
-        { role: "user", content: "What is Bitcoin?" },
+        { role: 'user', content: 'What is Bitcoin?' }
     ];
     const bobHistory = [
         {
-            role: "system",
-            content: "You ask questions about Bitcoin in one sentence.",
-        },
+            role: 'system',
+            content: 'You ask questions about Bitcoin in one sentence.'
+        }
     ];
-    console.log("Bob: ", "What is Bitcoin?");
+    console.log('Bob: ', 'What is Bitcoin?');
     const aliceResponse = completion({
         modelId: alice,
         history: [...aliceHistory],
-        stream: false,
+        stream: false
     });
     const aliceText = await aliceResponse.text;
-    aliceHistory.push({ role: "assistant", content: aliceText });
-    console.log("Alice: ", aliceText);
-    console.log("▸ Alice Stats: ", await aliceResponse.stats);
+    aliceHistory.push({ role: 'assistant', content: aliceText });
+    console.log('Alice: ', aliceText);
+    console.log('▸ Alice Stats: ', await aliceResponse.stats);
     const bobResponse = completion({
         modelId: bob,
-        history: [...bobHistory, { role: "user", content: aliceText }],
-        stream: false,
+        history: [...bobHistory, { role: 'user', content: aliceText }],
+        stream: false
     });
     const bobText = await bobResponse.text;
-    bobHistory.push({ role: "assistant", content: bobText });
-    console.log("Bob: ", bobText);
-    console.log("▸ Bob Stats: ", await bobResponse.stats);
+    bobHistory.push({ role: 'assistant', content: bobText });
+    console.log('Bob: ', bobText);
+    console.log('▸ Bob Stats: ', await bobResponse.stats);
     const aliceResponse2 = completion({
         modelId: alice,
-        history: [...aliceHistory, { role: "user", content: bobText }],
-        stream: false,
+        history: [...aliceHistory, { role: 'user', content: bobText }],
+        stream: false
     });
     const aliceText2 = await aliceResponse2.text;
-    aliceHistory.push({ role: "assistant", content: aliceText2 });
-    console.log("Alice: ", aliceText2);
-    console.log("▸ Alice Stats: ", await aliceResponse2.stats);
+    aliceHistory.push({ role: 'assistant', content: aliceText2 });
+    console.log('Alice: ', aliceText2);
+    console.log('▸ Alice Stats: ', await aliceResponse2.stats);
     const bobResponse2 = completion({
         modelId: bob,
-        history: [...bobHistory, { role: "user", content: aliceText2 }],
-        stream: false,
+        history: [...bobHistory, { role: 'user', content: aliceText2 }],
+        stream: false
     });
     const bobText2 = await bobResponse2.text;
-    bobHistory.push({ role: "assistant", content: bobText2 });
-    console.log("Bob: ", bobText2);
-    console.log("▸ Bob Stats: ", await bobResponse2.stats);
+    bobHistory.push({ role: 'assistant', content: bobText2 });
+    console.log('Bob: ', bobText2);
+    console.log('▸ Bob Stats: ', await bobResponse2.stats);
     await unloadModel({ modelId: alice, clearStorage: false });
     await unloadModel({ modelId: bob, clearStorage: false });
 }
 catch (error) {
-    console.error("✖", error);
+    console.error('✖', error);
     process.exit(1);
 }

package/dist/examples/ocr-fasttext.js

@@ -9,23 +9,23 @@
  * Pass a custom image path, or download the default image into examples/image/:
  *   https://github.com/tetherto/qvac/blob/main/packages/sdk/examples/image/basic_test.bmp
  */
-import { close, loadModel, ocr, OCR_LATIN, unloadModel, } from "@qvac/sdk";
-import path from "path";
-import { fileURLToPath } from "url";
+import { close, loadModel, ocr, OCR_LATIN, unloadModel } from '@qvac/sdk';
+import path from 'path';
+import { fileURLToPath } from 'url';
 const __dirname = path.dirname(fileURLToPath(import.meta.url));
-const imagePath = process.argv[2] || path.join(__dirname, "image/basic_test.bmp");
+const imagePath = process.argv[2] || path.join(__dirname, 'image/basic_test.bmp');
 try {
-    console.log("▸ Loading OCR model...");
+    console.log('▸ Loading OCR model...');
     const modelId = await loadModel({
         modelSrc: OCR_LATIN,
         modelConfig: {
-            langList: ["en"],
+            langList: ['en'],
             magRatio: 1.5,
             defaultRotationAngles: [90, 180, 270],
             contrastRetry: false,
             lowConfidenceThreshold: 0.5,
-            recognizerBatchSize: 1,
-        },
+            recognizerBatchSize: 1
+        }
     });
     console.log(`▸ Model loaded successfully! Model ID: ${modelId}`);
     console.log(`\n▸ Running OCR on: ${imagePath}`);
@@ -33,28 +33,28 @@ try {
         modelId,
         image: imagePath,
         options: {
-            paragraph: false,
-        },
+            paragraph: false
+        }
     });
     const result = await blocks;
-    console.log("\n▸ OCR Results:");
-    console.log("▸ ================================");
+    console.log('\n▸ OCR Results:');
+    console.log('▸ ================================');
     for (const block of result) {
         console.log(block.text);
         if (block.bbox) {
-            console.log(`▸ BBox: [${block.bbox.join(", ")}]`);
+            console.log(`▸ BBox: [${block.bbox.join(', ')}]`);
         }
         if (block.confidence !== undefined) {
             console.log(`▸ Confidence: ${block.confidence}`);
         }
     }
-    console.log("\n▸ ================================");
-    console.log("\n▸ Unloading model...");
+    console.log('\n▸ ================================');
+    console.log('\n▸ Unloading model...');
     await unloadModel({ modelId, clearStorage: false });
-    console.log("▸ Model unloaded successfully.");
+    console.log('▸ Model unloaded successfully.');
     process.exit(0);
 }
 catch (error) {
-    console.error("✖", error);
+    console.error('✖', error);
     await close();
 }

package/dist/examples/parallel-download.js

@@ -1,12 +1,12 @@
-import { LLAMA_3_2_1B_INST_Q4_0, WHISPER_TINY, VAD_SILERO_5_1_2, GTE_LARGE_FP16, downloadAsset, close, } from "@qvac/sdk";
+import { LLAMA_3_2_1B_INST_Q4_0, WHISPER_TINY, VAD_SILERO_5_1_2, GTE_LARGE_FP16, downloadAsset, close } from '@qvac/sdk';
 function now() {
     return Date.now();
 }
 const assets = [
-    { name: "Llama 3.2 1B", src: LLAMA_3_2_1B_INST_Q4_0 },
-    { name: "Whisper Tiny", src: WHISPER_TINY },
-    { name: "VAD Silero", src: VAD_SILERO_5_1_2 },
-    { name: "GTE Large FP16", src: GTE_LARGE_FP16 },
+    { name: 'Llama 3.2 1B', src: LLAMA_3_2_1B_INST_Q4_0 },
+    { name: 'Whisper Tiny', src: WHISPER_TINY },
+    { name: 'VAD Silero', src: VAD_SILERO_5_1_2 },
+    { name: 'GTE Large FP16', src: GTE_LARGE_FP16 }
 ];
 const timers = {};
 console.log(`\n▸ Parallel Download (${assets.length} assets)\n`);
@@ -26,7 +26,7 @@ try {
                     t.end = now();
                     console.log(`▸ [${asset.name}] done at ${((t.end - wallStart) / 1000).toFixed(1)}s`);
                 }
-            },
+            }
         });
     });
     const results = await Promise.allSettled(promises);
@@ -36,18 +36,16 @@ try {
         const asset = assets[i];
         const result = results[i];
         const t = timers[asset.name];
-        const status = result.status === "fulfilled" ? "OK" : "FAILED";
-        const reason = result.status === "rejected" ? ` — ${result.reason}` : "";
-        const timeToFirst = t.firstProgress != null
-            ? `${((t.firstProgress - t.start) / 1000).toFixed(1)}s`
-            : "N/A";
-        const total = t.end != null ? `${((t.end - t.start) / 1000).toFixed(1)}s` : "N/A";
+        const status = result.status === 'fulfilled' ? 'OK' : 'FAILED';
+        const reason = result.status === 'rejected' ? ` — ${result.reason}` : '';
+        const timeToFirst = t.firstProgress != null ? `${((t.firstProgress - t.start) / 1000).toFixed(1)}s` : 'N/A';
+        const total = t.end != null ? `${((t.end - t.start) / 1000).toFixed(1)}s` : 'N/A';
         console.log(`▸ ${status} ${asset.name}: first-progress=${timeToFirst}, total=${total}${reason}`);
     }
     console.log(`\n▸ Wall time: ${((wallEnd - wallStart) / 1000).toFixed(1)}s`);
     void close();
 }
 catch (error) {
-    console.error("✖", error);
+    console.error('✖', error);
     process.exit(1);
 }