Fixed: Now listens to model selection in job.json file

README.md

@@ -0,0 +1,527 @@
+# Big Link Man - Content Automation & Syndication Platform
+
+AI-powered content generation and multi-tier link building system with cloud deployment.
+
+## Quick Start
+
+```bash
+# Install dependencies
+uv pip install -r requirements.txt
+
+# Setup environment
+cp env.example .env
+# Edit .env with your credentials
+
+# Initialize database
+uv run python scripts/init_db.py
+
+# Create first admin user
+uv run python scripts/create_first_admin.py
+
+# Run CLI
+uv run python main.py --help
+```
+
+## Environment Configuration
+
+Required environment variables in `.env`:
+
+```bash
+DATABASE_URL=sqlite:///./content_automation.db
+OPENROUTER_API_KEY=your_key_here
+BUNNY_ACCOUNT_API_KEY=your_bunny_key_here
+```
+
+See `env.example` for full configuration options.
+
+## Database Management
+
+### Initialize Database
+```bash
+uv run python scripts/init_db.py
+```
+
+### Reset Database (drops all data)
+```bash
+uv run python scripts/init_db.py reset
+```
+
+### Create First Admin
+```bash
+uv run python scripts/create_first_admin.py
+```
+
+### Database Migrations
+```bash
+# Story 3.1 - Site deployments
+uv run python scripts/migrate_story_3.1_sqlite.py
+
+# Story 3.2 - Anchor text
+uv run python scripts/migrate_add_anchor_text.py
+
+# Story 3.3 - Template fields
+uv run python scripts/migrate_add_template_fields.py
+
+# Story 3.4 - Site pages
+uv run python scripts/migrate_add_site_pages.py
+
+# Story 4.1 - Deployment fields
+uv run python scripts/migrate_add_deployment_fields.py
+
+# Backfill site pages after migration
+uv run python scripts/backfill_site_pages.py
+```
+
+## User Management
+
+### Add User
+```bash
+uv run python main.py add-user \
+  --username newuser \
+  --password password123 \
+  --role Admin \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+### List Users
+```bash
+uv run python main.py list-users \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+### Delete User
+```bash
+uv run python main.py delete-user \
+  --username olduser \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+## Site Management
+
+### Provision New Site
+```bash
+uv run python main.py provision-site \
+  --name "My Site" \
+  --domain www.example.com \
+  --storage-name my-storage-zone \
+  --region DE \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+Regions: `DE`, `NY`, `LA`, `SG`, `SYD`
+
+### Attach Domain to Existing Storage
+```bash
+uv run python main.py attach-domain \
+  --name "Another Site" \
+  --domain www.another.com \
+  --storage-name my-storage-zone \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+### Sync Existing Bunny.net Sites
+```bash
+# Dry run
+uv run python main.py sync-sites \
+  --admin-user admin \
+  --dry-run
+
+# Actually import
+uv run python main.py sync-sites \
+  --admin-user admin
+```
+
+### List Sites
+```bash
+uv run python main.py list-sites \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+### Get Site Details
+```bash
+uv run python main.py get-site \
+  --domain www.example.com \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+### Remove Site
+```bash
+uv run python main.py remove-site \
+  --domain www.example.com \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+## Project Management
+
+### Ingest CORA Report
+```bash
+uv run python main.py ingest-cora \
+  --file shaft_machining.xlsx \
+  --name "Shaft Machining Project" \
+  --custom-anchors "shaft repair,engine parts" \
+  --username admin \
+  --password adminpass
+```
+
+### List Projects
+```bash
+uv run python main.py list-projects \
+  --username admin \
+  --password adminpass
+```
+
+## Content Generation
+
+### Create Job Configuration
+```bash
+# Tier 1 only
+uv run python create_job_config.py 1 tier1 15
+
+# Multi-tier
+uv run python create_job_config.py 1 multi 15 50 100
+```
+
+### Generate Content Batch
+```bash
+uv run python main.py generate-batch \
+  --job-file jobs/project_1_tier1_15articles.json \
+  --username admin \
+  --password adminpass
+```
+
+With options:
+```bash
+uv run python main.py generate-batch \
+  --job-file jobs/my_job.json \
+  --username admin \
+  --password adminpass \
+  --debug \
+  --continue-on-error \
+  --model gpt-4o-mini
+```
+
+Available models: `gpt-4o-mini`, `claude-sonnet-4.5`
+
+**Note:** If your job file contains a `models` config, it will override the `--model` flag and use different models for title, outline, and content generation stages.
+
+## Deployment
+
+### Deploy Batch
+```bash
+# Automatic deployment (runs after generation)
+uv run python main.py generate-batch \
+  --job-file jobs/my_job.json \
+  --username admin \
+  --password adminpass
+
+# Manual deployment
+uv run python main.py deploy-batch \
+  --batch-id 123 \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+### Dry Run Deployment
+```bash
+uv run python main.py deploy-batch \
+  --batch-id 123 \
+  --dry-run
+```
+
+### Verify Deployment
+```bash
+# Check all URLs
+uv run python main.py verify-deployment --batch-id 123
+
+# Check random sample
+uv run python main.py verify-deployment \
+  --batch-id 123 \
+  --sample 10 \
+  --timeout 10
+```
+
+## Link Export
+
+### Export Article URLs
+```bash
+# Tier 1 only
+uv run python main.py get-links \
+  --project-id 123 \
+  --tier 1
+
+# Tier 2 and above
+uv run python main.py get-links \
+  --project-id 123 \
+  --tier 2+
+
+# With anchor text and destinations
+uv run python main.py get-links \
+  --project-id 123 \
+  --tier 2+ \
+  --with-anchor-text \
+  --with-destination-url
+```
+
+Output is CSV format to stdout. Redirect to save:
+```bash
+uv run python main.py get-links \
+  --project-id 123 \
+  --tier 1 > tier1_urls.csv
+```
+
+## Utility Scripts
+
+### Check Last Generated Content
+```bash
+uv run python check_last_gen.py
+```
+
+### List All Users (Direct DB Access)
+```bash
+uv run python scripts/list_users.py
+```
+
+### Add Admin (Direct DB Access)
+```bash
+uv run python scripts/add_admin_direct.py
+```
+
+### Check Migration Status
+```bash
+uv run python scripts/check_migration.py
+```
+
+### Add Tier to Projects
+```bash
+uv run python scripts/add_tier_to_projects.py
+```
+
+## Testing
+
+### Run All Tests
+```bash
+uv run pytest
+```
+
+### Run Unit Tests
+```bash
+uv run pytest tests/unit/ -v
+```
+
+### Run Integration Tests
+```bash
+uv run pytest tests/integration/ -v
+```
+
+### Run Specific Test File
+```bash
+uv run pytest tests/unit/test_url_generator.py -v
+```
+
+### Run Story 3.1 Tests
+```bash
+uv run pytest tests/unit/test_url_generator.py \
+               tests/unit/test_site_provisioning.py \
+               tests/unit/test_site_assignment.py \
+               tests/unit/test_job_config_extensions.py \
+               tests/integration/test_story_3_1_integration.py \
+               -v
+```
+
+### Run with Coverage
+```bash
+uv run pytest --cov=src --cov-report=html
+```
+
+## System Information
+
+### Show Configuration
+```bash
+uv run python main.py config
+```
+
+### Health Check
+```bash
+uv run python main.py health
+```
+
+### List Available Models
+```bash
+uv run python main.py models
+```
+
+## Directory Structure
+
+```
+Big-Link-Man/
+├── main.py                 # CLI entry point
+├── src/                    # Source code
+│   ├── api/               # FastAPI endpoints
+│   ├── auth/              # Authentication
+│   ├── cli/               # CLI commands
+│   ├── core/              # Configuration
+│   ├── database/          # Models, repositories
+│   ├── deployment/        # Cloud deployment
+│   ├── generation/        # Content generation
+│   ├── ingestion/         # CORA parsing
+│   ├── interlinking/      # Link injection
+│   └── templating/        # HTML templates
+├── scripts/               # Database & utility scripts
+├── tests/                 # Test suite
+│   ├── unit/
+│   └── integration/
+├── jobs/                  # Job configuration files
+├── docs/                  # Documentation
+└── deployment_logs/       # Deployed URL logs
+```
+
+## Job Configuration Format
+
+Example job config (`jobs/example.json`):
+
+```json
+{
+  "job_name": "Multi-Tier Launch",
+  "project_id": 1,
+  "description": "Site build with 165 articles",
+  "models": {
+    "title": "openai/gpt-4o-mini",
+    "outline": "anthropic/claude-3.5-sonnet",
+    "content": "anthropic/claude-3.5-sonnet"
+  },
+  "tiers": [
+    {
+      "tier": 1,
+      "article_count": 15,
+      "validation_attempts": 3
+    },
+    {
+      "tier": 2,
+      "article_count": 50,
+      "validation_attempts": 2
+    }
+  ],
+  "failure_config": {
+    "max_consecutive_failures": 10,
+    "skip_on_failure": true
+  },
+  "interlinking": {
+    "links_per_article_min": 2,
+    "links_per_article_max": 4,
+    "include_home_link": true
+  },
+  "deployment_targets": ["www.primary.com"],
+  "tier1_preferred_sites": ["www.premium.com"],
+  "auto_create_sites": true
+}
+```
+
+### Per-Stage Model Configuration
+
+You can specify different AI models for each generation stage (title, outline, content):
+
+```json
+{
+  "models": {
+    "title": "openai/gpt-4o-mini",
+    "outline": "anthropic/claude-3.5-sonnet",
+    "content": "openai/gpt-4o"
+  }
+}
+```
+
+**Available models:**
+- `openai/gpt-4o-mini` - Fast and cost-effective
+- `openai/gpt-4o` - Higher quality, more expensive
+- `anthropic/claude-3.5-sonnet` - Excellent for long-form content
+
+If `models` is not specified in the job file, all stages use the model from the `--model` CLI flag (default: `gpt-4o-mini`).
+
+## Common Workflows
+
+### Initial Setup
+```bash
+uv pip install -r requirements.txt
+cp env.example .env
+# Edit .env
+uv run python scripts/init_db.py
+uv run python scripts/create_first_admin.py
+uv run python main.py sync-sites --admin-user admin
+```
+
+### New Project Workflow
+```bash
+# 1. Ingest CORA report
+uv run python main.py ingest-cora \
+  --file project.xlsx \
+  --name "My Project" \
+  --username admin \
+  --password adminpass
+
+# 2. Create job config
+uv run python create_job_config.py 1 multi 15 50 100
+
+# 3. Generate content (auto-deploys)
+uv run python main.py generate-batch \
+  --job-file jobs/project_1_multi_3tiers_165articles.json \
+  --username admin \
+  --password adminpass
+
+# 4. Verify deployment
+uv run python main.py verify-deployment --batch-id 1
+
+# 5. Export URLs for link building
+uv run python main.py get-links \
+  --project-id 1 \
+  --tier 1 > tier1_urls.csv
+```
+
+### Re-deploy After Changes
+```bash
+uv run python main.py deploy-batch \
+  --batch-id 123 \
+  --admin-user admin \
+  --admin-password adminpass
+```
+
+## Troubleshooting
+
+### Database locked
+```bash
+# Stop any running processes, then:
+uv run python scripts/init_db.py reset
+```
+
+### Missing dependencies
+```bash
+uv pip install -r requirements.txt --force-reinstall
+```
+
+### AI API errors
+Check `OPENROUTER_API_KEY` in `.env`
+
+### Bunny.net authentication failed
+Check `BUNNY_ACCOUNT_API_KEY` in `.env`
+
+### Storage upload failed
+Verify `storage_zone_password` in database (set during site provisioning)
+
+## Documentation
+
+- Product Requirements: `docs/prd.md`
+- Architecture: `docs/architecture/`
+- Implementation Summaries: `STORY_*.md` files
+- Quick Start Guides: `*_QUICKSTART.md` files
+
+## License
+
+All rights reserved.
+

docs/job-schema.md

@@ -35,7 +35,7 @@ Each job object defines a complete content generation batch for a specific proje
 
 | Field | Type | Default | Description |
 |-------|------|---------|-------------|
-| `models` | `Object` | Uses CLI default | AI models to use for each generation stage (Story 2.3 - planned) |
+| `models` | `Object` | Uses CLI default | AI models to use for each generation stage (title, outline, content) |
 | `deployment_targets` | `Array<string>` | `null` | Array of site custom_hostnames for tier1 deployment assignment (Story 2.5) |
 | `tier1_preferred_sites` | `Array<string>` | `null` | Array of hostnames for tier1 site assignment priority (Story 3.1) |
 | `auto_create_sites` | `boolean` | `false` | Whether to auto-create sites when pool is insufficient (Story 3.1) |
@@ -174,13 +174,13 @@ Each tier in the `tiers` object defines content generation parameters for that s
 }
 ```
 
-## AI Model Configuration (Story 2.3 - Not Yet Implemented)
+## AI Model Configuration
 
 ### `models`
 - **Type**: `Object` (optional)
 - **Purpose**: Specifies AI models to use for each generation stage
 - **Behavior**: Allows different models for title, outline, and content generation
-- **Note**: Currently not parsed by job config - uses CLI `--model` flag instead
+- **Note**: If not specified, all stages use the model from CLI `--model` flag (default: `gpt-4o-mini`)
 
 #### Models Object Fields
 | Field | Type | Description |
@@ -210,7 +210,7 @@ Each tier in the `tiers` object defines content generation parameters for that s
 ```
 
 ### Implementation Status
-This field is defined in the JSON schema but **not yet implemented** in the job config parser (`src/generation/job_config.py`). Currently, all stages use the same model specified via CLI `--model` flag.
+**Implemented** - The `models` field is fully functional. Different models can be specified for title, outline, and content generation stages. If a job file contains a `models` configuration and you also use the `--model` CLI flag, the system will warn you that the CLI flag is being ignored in favor of the job config.
 
 ## Tiered Link Configuration (Story 3.2)
 
@@ -299,7 +299,7 @@ This field is defined in the JSON schema but **not yet implemented** in the job
 ### Job Level Validation
 - `project_id` must be a positive integer
 - `tiers` must be an object with at least one tier
-- `models` must be an object with `title`, `outline`, and `content` fields (if specified) - **NOT YET VALIDATED**
+- `models` must be an object with `title`, `outline`, and `content` fields (if specified)
 - `deployment_targets` must be an array of strings (if specified)
 - `tier1_preferred_sites` must be an array of strings (if specified)
 - `auto_create_sites` must be a boolean (if specified)
@@ -331,7 +331,7 @@ uv run python main.py generate-batch --job-file jobs/example.json --username adm
 - `--password, -p`: Password for authentication
 - `--debug`: Save AI responses to debug_output/
 - `--continue-on-error`: Continue processing if article generation fails
-- `--model, -m`: AI model to use (default: gpt-4o-mini)
+- `--model, -m`: AI model to use (default: gpt-4o-mini). Overridden by job file `models` config if present.
 
 ## Implementation History
 
@@ -340,10 +340,11 @@ uv run python main.py generate-batch --job-file jobs/example.json --username adm
 - Added tier configuration with word count and heading constraints
 - Added tier defaults for common configurations
 
-### Story 2.3: AI Content Generation (Partial)
+### Story 2.3: AI Content Generation
-- **Implemented**: Database fields for tracking models (title_model, outline_model, content_model)
+- **Implemented**: Per-stage model selection via job config `models` field
-- **Not Implemented**: Job config `models` field - currently uses CLI `--model` flag
+- **Implemented**: Dynamic model switching in AIClient with `override_model` parameter
-- **Planned**: Per-stage model selection from job configuration
+- **Implemented**: CLI warning when job contains models but `--model` flag is used
+- **Behavior**: Job file `models` config takes precedence over CLI `--model` flag
 
 ### Story 2.5: Deployment Target Assignment
 - Added `deployment_targets` field for tier1 site assignment

scripts/backfill_site_pages.py

@@ -14,7 +14,7 @@ from src.database.session import db_manager
 from src.database.repositories import SiteDeploymentRepository, SitePageRepository, UserRepository
 from src.templating.service import TemplateService
 from src.generation.site_page_generator import generate_site_pages
-from src.auth.password import verify_password
+from src.auth.service import AuthService
 
 logging.basicConfig(
     level=logging.INFO,
@@ -44,7 +44,7 @@ def backfill_site_pages(
     user_repo = UserRepository(session)
     user = user_repo.get_by_username(username)
     
-    if not user or not verify_password(password, user.hashed_password):
+    if not user or not AuthService.verify_password(password, user.hashed_password):
         logger.error("Authentication failed")
         session.close()
         sys.exit(1)

src/cli/commands.py

@@ -932,7 +932,16 @@ def generate_batch(
                 click.echo("Please set OPENROUTER_API_KEY in your .env file", err=True)
                 raise click.Abort()
             
-            click.echo(f"Initializing AI client with model: {model}")
+            from src.generation.job_config import JobConfig
+            job_config = JobConfig(job_file)
+            jobs = job_config.get_jobs()
+            
+            has_models_in_job = any(job.models is not None for job in jobs)
+            if has_models_in_job and model != 'gpt-4o-mini':
+                click.echo(f"Warning: Job file contains per-stage model configuration.")
+                click.echo(f"         The --model flag will be ignored in favor of job config.\n")
+            
+            click.echo(f"Initializing AI client with default model: {model}")
             ai_client = AIClient(api_key=api_key, model=model)
             prompt_manager = PromptManager()
             

src/generation/ai_client.py

@@ -37,7 +37,8 @@ class AIClient:
         system_message: Optional[str] = None,
         max_tokens: int = 4000,
         temperature: float = 0.7,
-        json_mode: bool = False
+        json_mode: bool = False,
+        override_model: Optional[str] = None
     ) -> str:
         """
         Generate completion from OpenRouter API
@@ -48,6 +49,7 @@ class AIClient:
             max_tokens: Maximum tokens to generate
             temperature: Sampling temperature (0-1)
             json_mode: If True, requests JSON response format
+            override_model: If provided, use this model instead of self.model
         
         Returns:
             Generated text completion
@@ -57,8 +59,10 @@ class AIClient:
             messages.append({"role": "system", "content": system_message})
         messages.append({"role": "user", "content": prompt})
         
+        model_to_use = override_model if override_model else self.model
+        
         kwargs: Dict[str, Any] = {
-            "model": self.model,
+            "model": model_to_use,
             "messages": messages,
             "max_tokens": max_tokens,
             "temperature": temperature

src/generation/batch_processor.py

@@ -82,12 +82,20 @@ class BatchProcessor:
         auto_deploy: bool = True
     ):
         """Process a single job"""
+        self.current_job = job
+        
         project = self.project_repo.get_by_id(job.project_id)
         if not project:
             raise ValueError(f"Project {job.project_id} not found")
         
         click.echo(f"\nProcessing Job {job_idx}/{self.stats['total_jobs']}: Project ID {job.project_id}")
         
+        if job.models:
+            click.echo(f"  Using per-stage models:")
+            click.echo(f"    Title:   {job.models.title}")
+            click.echo(f"    Outline: {job.models.outline}")
+            click.echo(f"    Content: {job.models.content}")
+        
         resolved_targets = {}
         if job.deployment_targets:
             if not self.site_deployment_repo:
@@ -205,6 +213,8 @@ class BatchProcessor:
         """Generate a single article"""
         prefix = f"    [{article_num}/{tier_config.count}]"
         
+        models = self.current_job.models if hasattr(self, 'current_job') and self.current_job.models else None
+        
         site_deployment_id = assign_site_for_article(article_index, resolved_targets)
         
         if site_deployment_id:
@@ -214,7 +224,11 @@ class BatchProcessor:
             click.echo(f"{prefix} No site assignment (index {article_index} >= {len(resolved_targets)} targets)")
         
         click.echo(f"{prefix} Generating title...")
-        title = self.generator.generate_title(project_id, debug=debug)
+        title = self.generator.generate_title(
+            project_id, 
+            debug=debug,
+            model=models.title if models else None
+        )
         click.echo(f"{prefix} Generated title: \"{title}\"")
         
         click.echo(f"{prefix} Generating outline...")
@@ -225,7 +239,8 @@ class BatchProcessor:
             max_h2=tier_config.max_h2_tags,
             min_h3=tier_config.min_h3_tags,
             max_h3=tier_config.max_h3_tags,
-            debug=debug
+            debug=debug,
+            model=models.outline if models else None
         )
         
         h2_count = len(outline["outline"])
@@ -239,7 +254,8 @@ class BatchProcessor:
             outline=outline,
             min_word_count=tier_config.min_word_count,
             max_word_count=tier_config.max_word_count,
-            debug=debug
+            debug=debug,
+            model=models.content if models else None
         )
         
         word_count = self.generator.count_words(content)
@@ -253,7 +269,8 @@ class BatchProcessor:
                 content=content,
                 target_word_count=tier_config.min_word_count,
                 debug=debug,
-                project_id=project_id
+                project_id=project_id,
+                model=models.content if models else None
             )
             word_count = self.generator.count_words(content)
             click.echo(f"{prefix} Augmented content: {word_count:,} words")

src/generation/service.py

@@ -32,13 +32,14 @@ class ContentGenerator:
         self.template_service = template_service or TemplateService(content_repo)
         self.site_deployment_repo = site_deployment_repo
     
-    def generate_title(self, project_id: int, debug: bool = False) -> str:
+    def generate_title(self, project_id: int, debug: bool = False, model: Optional[str] = None) -> str:
         """
         Generate SEO-optimized title
         
         Args:
             project_id: Project ID to generate title for
             debug: If True, save response to debug_output/
+            model: Optional model override for this generation stage
         
         Returns:
             Generated title string
@@ -61,7 +62,8 @@ class ContentGenerator:
             prompt=user_prompt,
             system_message=system_msg,
             max_tokens=100,
-            temperature=0.7
+            temperature=0.7,
+            override_model=model
         )
         
         title = title.strip().strip('"').strip("'")
@@ -81,7 +83,8 @@ class ContentGenerator:
         max_h2: int,
         min_h3: int,
         max_h3: int,
-        debug: bool = False
+        debug: bool = False,
+        model: Optional[str] = None
     ) -> dict:
         """
         Generate article outline in JSON format
@@ -94,6 +97,7 @@ class ContentGenerator:
             min_h3: Minimum H3 subheadings total
             max_h3: Maximum H3 subheadings total
             debug: If True, save response to debug_output/
+            model: Optional model override for this generation stage
         
         Returns:
             Outline dictionary: {"outline": [{"h2": "...", "h3": ["...", "..."]}]}
@@ -125,7 +129,8 @@ class ContentGenerator:
             system_message=system_msg,
             max_tokens=2000,
             temperature=0.7,
-            json_mode=True
+            json_mode=True,
+            override_model=model
         )
         print(f"[DEBUG] Raw outline response: {outline_json}")
         # Save raw response immediately
@@ -168,7 +173,8 @@ class ContentGenerator:
         outline: dict,
         min_word_count: int,
         max_word_count: int,
-        debug: bool = False
+        debug: bool = False,
+        model: Optional[str] = None
     ) -> str:
         """
         Generate full article HTML fragment
@@ -180,6 +186,7 @@ class ContentGenerator:
             min_word_count: Minimum word count for guidance
             max_word_count: Maximum word count for guidance
             debug: If True, save response to debug_output/
+            model: Optional model override for this generation stage
         
         Returns:
             HTML string with <h2>, <h3>, <p> tags
@@ -207,7 +214,8 @@ class ContentGenerator:
             prompt=user_prompt,
             system_message=system_msg,
             max_tokens=8000,
-            temperature=0.7
+            temperature=0.7,
+            override_model=model
         )
         
         content = content.strip()
@@ -255,7 +263,8 @@ class ContentGenerator:
         content: str, 
         target_word_count: int,
         debug: bool = False,
-        project_id: Optional[int] = None
+        project_id: Optional[int] = None,
+        model: Optional[str] = None
     ) -> str:
         """
         Expand article content to meet minimum word count
@@ -265,6 +274,7 @@ class ContentGenerator:
             target_word_count: Target word count
             debug: If True, save response to debug_output/
             project_id: Optional project ID for debug output
+            model: Optional model override for this generation stage
         
         Returns:
             Expanded HTML content
@@ -279,7 +289,8 @@ class ContentGenerator:
             prompt=user_prompt,
             system_message=system_msg,
             max_tokens=8000,
-            temperature=0.7
+            temperature=0.7,
+            override_model=model
         )
         
         augmented = augmented.strip()