diff --git a/ml-model/js/cli.js b/ml-model/js/cli.js
index ae0fb47..e043deb 100755
--- a/ml-model/js/cli.js
+++ b/ml-model/js/cli.js
@@ -10,33 +10,45 @@ class OllamaAISelector {
 
     async getAvailableModels() {
         return new Promise((resolve, reject) => {
-            const ollama = spawn('ollama', ['list']);
-            let output = '';
-            let error = '';
+            try {
+                const ollama = spawn('ollama', ['list']);
+                let output = '';
+                let error = '';
 
-            ollama.stdout.on('data', (data) => {
-                output += data.toString();
-            });
+                ollama.stdout.on('data', (data) => {
+                    output += data.toString();
+                });
 
-            ollama.stderr.on('data', (data) => {
-                error += data.toString();
-            });
+                ollama.stderr.on('data', (data) => {
+                    error += data.toString();
+                });
 
-            ollama.on('close', (code) => {
-                if (code !== 0) {
-                    reject(new Error(`Ollama list failed: ${error}`));
-                    return;
-                }
+                ollama.on('close', (code) => {
+                    if (code !== 0) {
+                        reject(new Error(`Ollama list failed: ${error}`));
+                        return;
+                    }
 
-                // Parse ollama list output
-                const lines = output.trim().split('\n').slice(1); // Skip header
-                const models = lines
-                    .filter(line => line.trim())
-                    .map(line => line.split(/\s+/)[0])
-                    .filter(model => model && !model.includes('NAME'));
+                    // Parse ollama list output
+                    const lines = output.trim().split('\n').slice(1); // Skip header
+                    const models = lines
+                        .filter(line => line.trim())
+                        .map(line => line.split(/\s+/)[0])
+                        .filter(model => model && !model.includes('NAME'));
 
-                resolve(models);
-            });
+                    resolve(models);
+                });
+                
+                ollama.on('error', (err) => {
+                    if (err.code === 'ENOENT') {
+                        reject(new Error('Ollama not found. Please install Ollama from https://ollama.ai'));
+                    } else {
+                        reject(new Error(`Ollama spawn error: ${err.message}`));
+                    }
+                });
+            } catch (spawnError) {
+                reject(new Error(`Failed to start Ollama: ${spawnError.message}`));
+            }
         });
     }
 
@@ -44,14 +56,15 @@ class OllamaAISelector {
         console.log(`🚀 Running Ollama with ${modelName}...`);
         
         return new Promise((resolve, reject) => {
-            const args = ['run', modelName];
-            if (prompt) {
-                args.push(prompt);
-            }
+            try {
+                const args = ['run', modelName];
+                if (prompt) {
+                    args.push(prompt);
+                }
 
-            const ollama = spawn('ollama', args, { 
-                stdio: 'inherit' // Stream output directly to terminal
-            });
+                const ollama = spawn('ollama', args, { 
+                    stdio: 'inherit' // Stream output directly to terminal
+                });
 
             ollama.on('close', (code) => {
                 if (code === 0) {
@@ -61,9 +74,16 @@ class OllamaAISelector {
                 }
             });
 
-            ollama.on('error', (error) => {
-                reject(new Error(`Failed to start Ollama: ${error.message}`));
-            });
+                ollama.on('error', (error) => {
+                    if (error.code === 'ENOENT') {
+                        reject(new Error('Ollama not found. Please install Ollama from https://ollama.ai'));
+                    } else {
+                        reject(new Error(`Failed to start Ollama: ${error.message}`));
+                    }
+                });
+            } catch (spawnError) {
+                reject(new Error(`Failed to start Ollama: ${spawnError.message}`));
+            }
         });
     }
 
diff --git a/package.json b/package.json
index 4025617..7af982a 100644
--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
 {
   "name": "llm-checker",
-  "version": "2.3.2",
+  "version": "2.3.3",
   "description": "Intelligent CLI tool with AI-powered model selection that analyzes your hardware and recommends optimal LLM models for your system",
   "bin": {
     "llm-checker": "bin/enhanced_cli.js"
diff --git a/src/models/deterministic-selector.js b/src/models/deterministic-selector.js
index 3f0108c..4ef9ebe 100644
--- a/src/models/deterministic-selector.js
+++ b/src/models/deterministic-selector.js
@@ -221,7 +221,8 @@ class DeterministicModelSelector {
     }
 
     async getModelDetails(modelName) {
-        const details = await this.runOllamaCommand(['show', modelName]);
+        try {
+            const details = await this.runOllamaCommand(['show', modelName]);
         
         // Parse model details from ollama show output
         const meta = {
@@ -236,7 +237,22 @@ class DeterministicModelSelector {
             model_identifier: modelName
         };
         
-        return meta;
+            return meta;
+        } catch (error) {
+            // If Ollama is not available or model details can't be fetched, return minimal info
+            return {
+                name: modelName,
+                family: 'unknown',
+                paramsB: 0,
+                ctxMax: 2048,
+                quant: 'unknown',
+                sizeGB: 0,
+                modalities: ['text'],
+                tags: [],
+                model_identifier: modelName,
+                error: error.message
+            };
+        }
     }
 
     /**
@@ -404,20 +420,34 @@ class DeterministicModelSelector {
 
     async runOllamaCommand(args) {
         return new Promise((resolve, reject) => {
-            const proc = spawn('ollama', args, { stdio: 'pipe' });
-            let output = '';
-            let error = '';
-            
-            proc.stdout.on('data', (data) => output += data);
-            proc.stderr.on('data', (data) => error += data);
-            
-            proc.on('close', (code) => {
-                if (code === 0) {
-                    resolve(output);
-                } else {
-                    reject(new Error(`Ollama command failed: ${error}`));
-                }
-            });
+            try {
+                const proc = spawn('ollama', args, { stdio: 'pipe' });
+                let output = '';
+                let error = '';
+                
+                proc.stdout.on('data', (data) => output += data);
+                proc.stderr.on('data', (data) => error += data);
+                
+                proc.on('close', (code) => {
+                    if (code === 0) {
+                        resolve(output);
+                    } else {
+                        reject(new Error(`Ollama command failed: ${error}`));
+                    }
+                });
+                
+                proc.on('error', (err) => {
+                    // Handle ENOENT and other spawn errors gracefully
+                    if (err.code === 'ENOENT') {
+                        reject(new Error('Ollama not found. Please install Ollama from https://ollama.ai'));
+                    } else {
+                        reject(new Error(`Ollama spawn error: ${err.message}`));
+                    }
+                });
+            } catch (spawnError) {
+                // Handle synchronous spawn errors
+                reject(new Error(`Failed to start Ollama: ${spawnError.message}`));
+            }
         });
     }