@auxot/worker-cli 0.1.0 → 0.1.3

package/dist/model-downloader.js

@@ -1,145 +0,0 @@
-/**
- * Model Downloader
- *
- * Downloads GGUF model files from Hugging Face.
- *
- * Features:
- * - Progress reporting
- * - Resumable downloads (HTTP Range requests)
- * - Integrity verification (file size)
- * - Caching (checks if file exists)
- */
-import { createWriteStream, existsSync, statSync } from 'node:fs';
-import { mkdir } from 'node:fs/promises';
-import { dirname } from 'node:path';
-/**
- * Download a model file from Hugging Face
- *
- * @param entry Model registry entry
- * @param outputPath Full path where the file should be saved
- * @param onProgress Optional progress callback (bytes downloaded, total bytes)
- * @returns Path to downloaded file
- */
-export async function downloadModel(entry, outputPath, onProgress) {
-    // Create output directory if it doesn't exist
-    const outputDir = dirname(outputPath);
-    if (!existsSync(outputDir)) {
-        await mkdir(outputDir, { recursive: true });
-    }
-    // Check if file already exists
-    if (existsSync(outputPath)) {
-        const stats = statSync(outputPath);
-        // If file size matches expected size, skip download
-        if (entry.file_size_bytes && stats.size === entry.file_size_bytes) {
-            console.log(`  ✓ Model already downloaded (${formatBytes(stats.size)})`);
-            return outputPath;
-        }
-        // If file exists but size doesn't match, delete it and re-download
-        if (entry.file_size_bytes && stats.size !== entry.file_size_bytes) {
-            console.log(`  ⊘ Existing file size mismatch (${formatBytes(stats.size)} vs ${formatBytes(entry.file_size_bytes)})`);
-            console.log(`  ⊘ Re-downloading...`);
-            // Delete the file - we'll download fresh
-            const { unlink } = await import('node:fs/promises');
-            await unlink(outputPath);
-        }
-    }
-    // Build Hugging Face download URL
-    // Format: https://huggingface.co/{repo_id}/resolve/main/{filename}
-    const downloadUrl = `https://huggingface.co/${entry.huggingface_id}/resolve/main/${entry.file_name}`;
-    console.log(`  Downloading from: ${entry.huggingface_id}`);
-    console.log(`  File: ${entry.file_name}`);
-    if (entry.file_size_bytes) {
-        console.log(`  Size: ${formatBytes(entry.file_size_bytes)}`);
-    }
-    // Download with progress tracking
-    let downloadedBytes = 0;
-    const totalBytes = entry.file_size_bytes || 0;
-    // Check if we can resume (partial file exists)
-    let startByte = 0;
-    if (existsSync(outputPath)) {
-        const stats = statSync(outputPath);
-        startByte = stats.size;
-        if (startByte > 0 && startByte < totalBytes) {
-            console.log(`  Resuming from ${formatBytes(startByte)}...`);
-            downloadedBytes = startByte;
-        }
-    }
-    const response = await fetch(downloadUrl, {
-        headers: startByte > 0 ? {
-            'Range': `bytes=${startByte}-`,
-        } : {},
-    });
-    if (!response.ok) {
-        if (response.status === 416) {
-            // Range not satisfiable - file already fully downloaded
-            if (existsSync(outputPath)) {
-                console.log(`  ✓ Download complete`);
-                return outputPath;
-            }
-        }
-        throw new Error(`Download failed: ${response.status} ${response.statusText}`);
-    }
-    const contentLength = response.headers.get('content-length');
-    const totalSize = contentLength ? parseInt(contentLength, 10) + startByte : totalBytes;
-    // Open file for writing (append if resuming)
-    const fileStream = createWriteStream(outputPath, { flags: startByte > 0 ? 'a' : 'w' });
-    // Stream response to file
-    const reader = response.body?.getReader();
-    if (!reader) {
-        throw new Error('Response body is not readable');
-    }
-    try {
-        while (true) {
-            const { done, value } = await reader.read();
-            if (done) {
-                break;
-            }
-            fileStream.write(value);
-            downloadedBytes += value.length;
-            // Report progress
-            if (onProgress) {
-                onProgress(downloadedBytes, totalSize);
-            }
-            else if (totalSize > 0) {
-                // Simple progress log every 10MB
-                if (downloadedBytes % (10 * 1024 * 1024) < value.length) {
-                    const percent = ((downloadedBytes / totalSize) * 100).toFixed(1);
-                    process.stdout.write(`\r  Progress: ${percent}% (${formatBytes(downloadedBytes)} / ${formatBytes(totalSize)})`);
-                }
-            }
-        }
-        fileStream.end();
-        // Wait for file stream to finish
-        await new Promise((resolve, reject) => {
-            fileStream.on('finish', resolve);
-            fileStream.on('error', reject);
-        });
-        if (totalSize > 0 && downloadedBytes !== totalSize) {
-            throw new Error(`Download incomplete: ${downloadedBytes} bytes downloaded, expected ${totalSize}`);
-        }
-        if (onProgress) {
-            // Clear progress line
-            process.stdout.write('\r');
-        }
-        else {
-            process.stdout.write('\r');
-        }
-        console.log(`  ✓ Download complete (${formatBytes(downloadedBytes)})`);
-        return outputPath;
-    }
-    catch (error) {
-        fileStream.destroy();
-        throw error;
-    }
-}
-/**
- * Format bytes to human-readable string
- */
-function formatBytes(bytes) {
-    if (bytes === 0)
-        return '0 B';
-    const k = 1024;
-    const sizes = ['B', 'KB', 'MB', 'GB', 'TB'];
-    const i = Math.floor(Math.log(bytes) / Math.log(k));
-    return `${(bytes / Math.pow(k, i)).toFixed(1)} ${sizes[i]}`;
-}

package/dist/model-resolver.js

@@ -1,80 +0,0 @@
-/**
- * Model Path Resolver
- *
- * Resolves model paths from policy using the model registry.
- * Downloads models if not cached.
- */
-import { loadRegistry, getModels } from '@auxot/model-registry';
-import { join } from 'path';
-import { homedir } from 'os';
-import { downloadModel } from './model-downloader.js';
-import { existsSync } from 'node:fs';
-/**
- * Get model entry from registry based on policy
- */
-export function getModelFromPolicy(policy) {
-    const registry = loadRegistry();
-    // Find model matching policy (model_name and quantization)
-    const matchingModels = getModels(registry, {
-        model_name: policy.model_name,
-    });
-    // Filter by quantization (case-insensitive)
-    const model = matchingModels.find((m) => m.quantization.toLowerCase() === policy.quantization.toLowerCase());
-    return model || null;
-}
-/**
- * Get model path from policy, downloading if necessary
- *
- * Returns the local path for the model file.
- * Downloads the model if it doesn't exist or is incomplete.
- *
- * @param policy GPU key policy
- * @param onProgress Optional progress callback (downloaded, total)
- * @returns Model file path, or null if model not found in registry
- */
-export async function ensureModelDownloaded(policy, onProgress) {
-    const model = getModelFromPolicy(policy);
-    if (!model) {
-        console.error(`  ✗ Model not found in registry: ${policy.model_name} (${policy.quantization})`);
-        return null;
-    }
-    // Build model path
-    const modelsDir = process.env.AUXOT_MODELS_DIR || join(homedir(), '.auxot', 'models');
-    const modelDir = join(modelsDir, model.huggingface_id.replace('/', '_'));
-    const modelPath = join(modelDir, model.file_name);
-    // Check if model exists and is valid
-    if (existsSync(modelPath)) {
-        const { statSync } = await import('node:fs');
-        const stats = statSync(modelPath);
-        // If file size matches expected size, use cached model
-        if (model.file_size_bytes && stats.size === model.file_size_bytes) {
-            return modelPath;
-        }
-    }
-    // Download model
-    console.log(`  Downloading model: ${model.model_name} (${model.quantization})`);
-    try {
-        await downloadModel(model, modelPath, onProgress);
-        return modelPath;
-    }
-    catch (error) {
-        console.error(`  ✗ Download failed:`, error);
-        throw error;
-    }
-}
-/**
- * Get model path without downloading (for checking if model exists)
- *
- * @param policy GPU key policy
- * @returns Model file path, or null if model not found in registry
- */
-export function getModelPath(policy) {
-    const model = getModelFromPolicy(policy);
-    if (!model) {
-        return null;
-    }
-    const modelsDir = process.env.AUXOT_MODELS_DIR || join(homedir(), '.auxot', 'models');
-    const modelDir = join(modelsDir, model.huggingface_id.replace('/', '_'));
-    const modelPath = join(modelDir, model.file_name);
-    return modelPath;
-}

package/dist/policy-validator.js

@@ -1,242 +0,0 @@
-/**
- * Policy Validation
- *
- * Validates worker capabilities against GPU key policy.
- */
-/**
- * Normalize model name for comparison
- * Extracts base model name + version + variant (removes quantization, parameters, file patterns)
- */
-function normalizeModelName(name) {
-    // Remove common quantization suffixes
-    let normalized = name
-        .replace(/-GGUF$/i, '')
-        .replace(/\.gguf$/i, '');
-    // Remove path components (keep only filename)
-    const parts = normalized.split('/');
-    if (parts.length > 1) {
-        normalized = parts[parts.length - 1];
-    }
-    // Remove multi-GGUF file patterns (e.g., "-00001-of-00003")
-    normalized = normalized.replace(/-\d{5}-of-\d{5}$/i, '');
-    // Extract version number (e.g., "3", "2.5", "4", "3.3")
-    let version = '';
-    const versionMatch = normalized.match(/^(Qwen|Llama|Ministral|Devstral|Gemma|DeepSeek|Granite|GPT-OSS)[-_]?(\d+(?:\.\d+)?)/i);
-    if (versionMatch) {
-        version = versionMatch[2];
-        normalized = normalized.replace(new RegExp(`^${versionMatch[1]}[-_]?${versionMatch[2]}`, 'i'), versionMatch[1]);
-    }
-    // Extract variant (composite or single)
-    let variant = '';
-    const compositePatterns = [
-        /-VL-Instruct$/i,
-        /-VL-Thinking$/i,
-        /-VL-Chat$/i,
-        /-VL-Coder$/i,
-        /-VL-Code$/i,
-    ];
-    for (const pattern of compositePatterns) {
-        const match = normalized.match(pattern);
-        if (match) {
-            variant = match[0].replace(/^-/i, '');
-            normalized = normalized.replace(pattern, '');
-            break;
-        }
-    }
-    if (!variant) {
-        const singlePatterns = [
-            /-Instruct$/i,
-            /-Thinking$/i,
-            /-Chat$/i,
-            /-Coder$/i,
-            /-Code$/i,
-            /-VL$/i,
-            /-Vision$/i,
-            /-Maverick$/i,
-            /-Scout$/i,
-            /-Reasoning$/i,
-        ];
-        for (const pattern of singlePatterns) {
-            const match = normalized.match(pattern);
-            if (match) {
-                variant = match[0].replace(/^-/i, '');
-                normalized = normalized.replace(pattern, '');
-                break;
-            }
-        }
-    }
-    // Remove quantization patterns (Q4_K_M, Q5_K_S, etc.)
-    normalized = normalized.replace(/[-_]Q\d+[_\w]*/i, '');
-    // Remove parameter counts (7B, 13B, 30B, etc.)
-    normalized = normalized.replace(/[-_](\d+(?:\.\d+)?[BMK])(?![0-9])/i, '');
-    // Remove expert counts (A22B, E2B, etc.)
-    normalized = normalized.replace(/[-_]([AE]\d+[BMK])/i, '');
-    normalized = normalized.replace(/[-_](\d+[AE])(?![0-9])/i, '');
-    // Remove trailing version numbers that are NOT part of the model name
-    normalized = normalized.replace(/[-_](\d{4,})$/i, ''); // Remove 4+ digit trailing numbers
-    // Normalize base name
-    const nameLower = normalized.toLowerCase();
-    let base = '';
-    if (nameLower.startsWith('qwen')) {
-        base = 'Qwen';
-    }
-    else if (nameLower.startsWith('llama') || nameLower.startsWith('meta-llama')) {
-        base = 'Llama';
-    }
-    else if (nameLower.startsWith('ministral') || nameLower.startsWith('devstral')) {
-        base = 'Ministral';
-    }
-    else if (nameLower.startsWith('gemma')) {
-        base = 'Gemma';
-    }
-    else if (nameLower.startsWith('deepseek')) {
-        base = 'DeepSeek';
-    }
-    else if (nameLower.startsWith('granite')) {
-        base = 'Granite';
-    }
-    else if (nameLower.startsWith('gpt-oss') || nameLower.startsWith('gptoss')) {
-        base = 'GPT-OSS';
-    }
-    else {
-        base = normalized.trim();
-    }
-    // Combine base + version + variant
-    let result = base;
-    if (version) {
-        result += ` ${version}`;
-    }
-    if (variant) {
-        result += `-${variant}`;
-    }
-    return result.trim();
-}
-/**
- * Extract quantization from model name or capabilities
- */
-function extractQuantization(capabilities) {
-    // Try to extract from model name
-    const model = capabilities.model || '';
-    // Check for quantization patterns
-    const quantPatterns = [
-        'Q3_K_S', 'Q4_K_S', 'Q5_K_S', 'Q6_K', 'Q8_0', 'Q8_K',
-        'F16', 'F32', 'BF16',
-    ];
-    for (const pattern of quantPatterns) {
-        if (model.includes(pattern)) {
-            return pattern;
-        }
-    }
-    return null;
-}
-/**
- * Infer capabilities from model name
- */
-function inferCapabilitiesFromModel(modelName) {
-    const name = modelName.toLowerCase();
-    const capabilities = [];
-    if (name.includes('vision') || name.includes('multimodal') || name.includes('vl-')) {
-        capabilities.push('vision');
-    }
-    if (name.includes('code') || name.includes('coder') || name.includes('starcoder')) {
-        capabilities.push('code');
-    }
-    if (name.includes('embed') || name.includes('embedding')) {
-        capabilities.push('embedding');
-    }
-    if (capabilities.length === 0) {
-        capabilities.push('chat');
-    }
-    return [...new Set(capabilities)];
-}
-/**
- * Validate worker capabilities against policy
- */
-/**
- * Parse parameter count from string (e.g., "7B" -> 7e9, "30B" -> 30e9)
- */
-function parseParameters(parameters) {
-    const match = parameters.match(/^(\d+(?:\.\d+)?)(B|M|K)$/i);
-    if (!match)
-        return 0;
-    const value = parseFloat(match[1]);
-    const unit = match[2].toUpperCase();
-    if (unit === 'B')
-        return value * 1e9;
-    if (unit === 'M')
-        return value * 1e6;
-    if (unit === 'K')
-        return value * 1e3;
-    return value;
-}
-export function validatePolicy(discoveredCapabilities, policy) {
-    const errors = [];
-    const warnings = []; // Declared but currently unused (reserved for future validation warnings)
-    // 0. Check model size limit for CPU mode (warning only - binary download already handles this)
-    // Note: This is just a warning since the policy comes from the server
-    // The actual binary selection (GPU vs CPU) happens in llama-binary.ts
-    // TODO: Add warnings here if needed in the future
-    // 1. Model name match (normalized comparison)
-    // Normalize both discovered and policy model names to base + version + variant
-    const discoveredNormalized = normalizeModelName(discoveredCapabilities.model || '');
-    const policyNormalized = normalizeModelName(policy.model_name);
-    if (discoveredNormalized !== policyNormalized) {
-        errors.push(`Model name mismatch: discovered "${discoveredCapabilities.model}" (normalized: "${discoveredNormalized}") ` +
-            `does not match policy "${policy.model_name}" (normalized: "${policyNormalized}")`);
-    }
-    // 2. Context size >= policy.context_size
-    const discoveredCtxSize = discoveredCapabilities.ctx_size || 0;
-    if (discoveredCtxSize < policy.context_size) {
-        errors.push(`Context size insufficient: discovered ${discoveredCtxSize} < required ${policy.context_size}`);
-    }
-    // 3. Quantization match (if specified in policy)
-    const discoveredQuant = extractQuantization(discoveredCapabilities);
-    if (discoveredQuant && discoveredQuant !== policy.quantization) {
-        errors.push(`Quantization mismatch: discovered "${discoveredQuant}" does not match policy "${policy.quantization}"`);
-    }
-    // 4. Capabilities match (worker must have all required capabilities)
-    const discoveredCaps = inferCapabilitiesFromModel(discoveredCapabilities.model || '');
-    const missingCaps = policy.capabilities.filter((requiredCap) => !discoveredCaps.includes(requiredCap));
-    if (missingCaps.length > 0) {
-        errors.push(`Missing required capabilities: ${missingCaps.join(', ')}. ` +
-            `Discovered: ${discoveredCaps.join(', ')}. ` +
-            `Required: ${policy.capabilities.join(', ')}`);
-    }
-    // 5. Parameters match (if specified in policy)
-    if (policy.parameters) {
-        const discoveredParams = discoveredCapabilities.parameters;
-        if (discoveredParams && discoveredParams !== policy.parameters) {
-            errors.push(`Parameters mismatch: discovered "${discoveredParams}" does not match policy "${policy.parameters}"`);
-        }
-    }
-    // 6. Family match (if specified in policy)
-    if (policy.family) {
-        // Infer family from model name or capabilities
-        // MoE models have:
-        // 1. "MoE" or "mixture-of-experts" in name
-        // 2. Expert notation like "A22B", "E22B" (e.g., "235B-A22B" = 235B total, 22B experts)
-        // 3. Multiple parameter counts separated by dashes (total-expert pattern)
-        const modelName = (discoveredCapabilities.model || '').toLowerCase();
-        // Check for explicit MoE indicators
-        const hasMoEKeyword = modelName.includes('moe') || modelName.includes('mixture-of-experts');
-        // Check for expert notation (A22B, E22B, etc.)
-        const hasExpertNotation = /[ae]\d+[bmk]/i.test(modelName);
-        // Check for total-expert parameter pattern (e.g., "235B-A22B", "70B-E2B")
-        const hasTotalExpertPattern = /\d+[bmk]-[ae]\d+[bmk]/i.test(modelName);
-        // Check for multiple large parameter counts separated by dashes
-        const paramPatterns = modelName.match(/\d+[bmk]/gi) || [];
-        const hasMultipleLargeParams = paramPatterns.length >= 2 &&
-            paramPatterns.some(p => /^(\d{2,}|[0-9]+0)[bmk]$/i.test(p)); // 2+ digits or ends in 0
-        const isMoE = hasMoEKeyword || hasExpertNotation || hasTotalExpertPattern || hasMultipleLargeParams;
-        const discoveredFamily = isMoE ? 'MoE' : 'Dense';
-        if (discoveredFamily !== policy.family) {
-            errors.push(`Family mismatch: discovered "${discoveredFamily}" does not match policy "${policy.family}". ` +
-                `Model name: "${discoveredCapabilities.model}"`);
-        }
-    }
-    return {
-        valid: errors.length === 0,
-        errors,
-        warnings: warnings.length > 0 ? warnings : undefined,
-    };
-}

package/dist/types.js

@@ -1,4 +0,0 @@
-/**
- * Type definitions for worker CLI
- */
-export {};