Text Generation
Transformers
English
qwen2
code-generation
python
fine-tuning
Qwen
tools
agent-framework
multi-agent
conversational
Eval Results (legacy)
Instructions to use my-ai-stack/Stack-2-9-finetuned with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use my-ai-stack/Stack-2-9-finetuned with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="my-ai-stack/Stack-2-9-finetuned") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("my-ai-stack/Stack-2-9-finetuned") model = AutoModelForCausalLM.from_pretrained("my-ai-stack/Stack-2-9-finetuned") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use my-ai-stack/Stack-2-9-finetuned with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "my-ai-stack/Stack-2-9-finetuned" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "my-ai-stack/Stack-2-9-finetuned", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/my-ai-stack/Stack-2-9-finetuned
- SGLang
How to use my-ai-stack/Stack-2-9-finetuned with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "my-ai-stack/Stack-2-9-finetuned" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "my-ai-stack/Stack-2-9-finetuned", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "my-ai-stack/Stack-2-9-finetuned" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "my-ai-stack/Stack-2-9-finetuned", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use my-ai-stack/Stack-2-9-finetuned with Docker Model Runner:
docker model run hf.co/my-ai-stack/Stack-2-9-finetuned
| #!/usr/bin/env python3 | |
| """ | |
| Integration Tests for Stack 2.9 Self-Evolution | |
| Self-improvement cycle tests. | |
| """ | |
| import pytest | |
| import sys | |
| from pathlib import Path | |
| from unittest.mock import MagicMock, patch, AsyncMock | |
| # Add stack_cli to path | |
| sys.path.insert(0, str(Path(__file__).parent.parent / "stack_cli")) | |
| from stack_cli.agent import StackAgent, SelfReflection, create_agent, AgentResponse | |
| class TestSelfReflection: | |
| """Test self-reflection functionality.""" | |
| def test_reflection_high_confidence(self): | |
| """Test reflection with high confidence.""" | |
| sr = SelfReflection() | |
| tool_calls = [ | |
| MagicMock(success=True, tool_name="read"), | |
| MagicMock(success=True, tool_name="write") | |
| ] | |
| result = sr.reflect("test query", tool_calls, "Good response with content") | |
| assert result["needs_reflection"] is False | |
| assert result["confidence"] >= 0.7 | |
| def test_reflection_low_confidence(self): | |
| """Test reflection with failures.""" | |
| sr = SelfReflection() | |
| tool_calls = [ | |
| MagicMock(success=False, tool_name="read", error="Not found") | |
| ] | |
| result = sr.reflect("test query", tool_calls, "Short") | |
| assert result["needs_reflection"] is True | |
| assert result["failed_calls"] > 0 | |
| def test_reflection_suggestion(self): | |
| """Test reflection provides suggestions.""" | |
| sr = SelfReflection() | |
| tool_calls = [ | |
| MagicMock(success=False, tool_name="read", error="Failed") | |
| ] | |
| result = sr.reflect("test query", tool_calls, "Short") | |
| assert result.get("suggestion") is not None | |
| class TestSelfImprovementCycle: | |
| """Test self-improvement cycle.""" | |
| def test_agent_learns_from_errors(self): | |
| """Test agent learns from errors.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| with patch('stack_cli.tools.get_tool') as mock_get_tool: | |
| # First call fails | |
| call_count = [0] | |
| def tool_func(**kwargs): | |
| call_count[0] += 1 | |
| if call_count[0] == 1: | |
| return {"success": False, "error": "First attempt failed"} | |
| return {"success": True} | |
| mock_get_tool.return_value = tool_func | |
| agent = StackAgent() | |
| # First attempt | |
| response1 = agent.process("read test.py") | |
| # Should have some response even on failure | |
| assert response1 is not None | |
| def test_conversation_history_tracking(self): | |
| """Test conversation history is tracked.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| with patch('stack_cli.tools.get_tool') as mock_get_tool: | |
| mock_tool = MagicMock(return_value={"success": True}) | |
| mock_get_tool.return_value = mock_tool | |
| agent = StackAgent() | |
| # Process multiple queries | |
| agent.process("query 1") | |
| agent.process("query 2") | |
| agent.process("query 3") | |
| assert len(agent.conversation_history) == 3 | |
| def test_reflection_updates_confidence(self): | |
| """Test reflection updates confidence score.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| with patch('stack_cli.tools.get_tool') as mock_get_tool: | |
| mock_tool = MagicMock(return_value={"success": True}) | |
| mock_get_tool.return_value = mock_tool | |
| agent = StackAgent() | |
| response = agent.process("test query") | |
| # Confidence should be set | |
| assert response.confidence is not None | |
| assert 0.0 <= response.confidence <= 1.0 | |
| class TestAdaptiveToolSelection: | |
| """Test adaptive tool selection.""" | |
| def test_tool_selection_based_on_intent(self): | |
| """Test tools are selected based on intent.""" | |
| from stack_cli.agent import ToolSelector | |
| ts = ToolSelector() | |
| # Different intents should select different tools | |
| tools_file = ts.select("file_read", {}) | |
| tools_git = ts.select("git_operation", {}) | |
| tools_web = ts.select("web_search", {}) | |
| assert tools_file != tools_git | |
| assert tools_git != tools_web | |
| def test_parameter_extraction_improves(self): | |
| """Test parameter extraction works across queries.""" | |
| from stack_cli.agent import ToolSelector | |
| ts = ToolSelector() | |
| # Same tool with different queries | |
| params1 = ts.get_tool_parameters("read", "read test.py", {}) | |
| params2 = ts.get_tool_parameters("read", "view main.py", {}) | |
| # Both should extract path | |
| assert "path" in params1 or "path" in params2 | |
| class TestContextAwareImprovement: | |
| """Test context-aware improvements.""" | |
| def test_context_influences_response(self): | |
| """Test context influences response generation.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| with patch('stack_cli.tools.get_tool') as mock_get_tool: | |
| mock_tool = MagicMock(return_value={"success": True, "content": "data"}) | |
| mock_get_tool.return_value = mock_tool | |
| agent = StackAgent() | |
| response1 = agent.process("read test.py") | |
| response2 = agent.process("read test.py") | |
| # Responses should be consistent | |
| assert response1.content is not None | |
| def test_session_memory_persists(self): | |
| """Test session memory persists across queries.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| with patch('stack_cli.tools.get_tool') as mock_get_tool: | |
| mock_tool = MagicMock(return_value={"success": True}) | |
| mock_get_tool.return_value = mock_tool | |
| agent = StackAgent() | |
| # Process query | |
| agent.process("first query") | |
| # Session should have recorded tool usage | |
| session = agent.context_manager.session | |
| assert session is not None | |
| class TestSelfEvolutionIntegration: | |
| """Test complete self-evolution integration.""" | |
| def test_full_self_improvement_loop(self): | |
| """Test complete self-improvement loop.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| with patch('stack_cli.tools.get_tool') as mock_get_tool: | |
| mock_tool = MagicMock(return_value={"success": True}) | |
| mock_get_tool.return_value = mock_tool | |
| agent = StackAgent() | |
| # 1. Process query | |
| response = agent.process("test query") | |
| # 2. Check reflection | |
| assert response is not None | |
| # 3. History is updated | |
| assert len(agent.conversation_history) > 0 | |
| def test_error_recovery_improves(self): | |
| """Test error recovery improves over time.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| with patch('stack_cli.tools.get_tool') as mock_get_tool: | |
| error_count = [0] | |
| def tool_func(**kwargs): | |
| error_count[0] += 1 | |
| if error_count[0] <= 2: | |
| return {"success": False, "error": f"Error {error_count[0]}"} | |
| return {"success": True} | |
| mock_get_tool.return_value = tool_func | |
| agent = StackAgent() | |
| # First attempts may fail | |
| for _ in range(3): | |
| agent.process("test") | |
| # Should have recorded history | |
| assert len(agent.conversation_history) >= 0 | |
| def test_performance_tracking(self): | |
| """Test performance is tracked.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| agent = StackAgent() | |
| # Get session summary | |
| summary = agent.context_manager.session.get_summary() | |
| assert "messages_count" in summary | |
| assert "tools_used_count" in summary | |
| class TestContinuousLearning: | |
| """Test continuous learning aspects.""" | |
| def test_query_patterns_learned(self): | |
| """Test query patterns are tracked.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| with patch('stack_cli.tools.get_tool') as mock_get_tool: | |
| mock_tool = MagicMock(return_value={"success": True}) | |
| mock_get_tool.return_value = mock_tool | |
| agent = StackAgent() | |
| # Similar queries | |
| agent.process("read file1.py") | |
| agent.process("read file2.py") | |
| agent.process("read file3.py") | |
| # History shows pattern | |
| history = agent.conversation_history | |
| assert len(history) == 3 | |
| def test_tool_usage_stats(self): | |
| """Test tool usage statistics.""" | |
| with patch('stack_cli.context.create_context_manager'): | |
| with patch('stack_cli.tools.get_tool') as mock_get_tool: | |
| mock_tool = MagicMock(return_value={"success": True}) | |
| mock_get_tool.return_value = mock_tool | |
| agent = StackAgent() | |
| # Use various tools | |
| agent.process("read test.py") | |
| agent.process("run pytest") | |
| agent.process("git status") | |
| # Session tracks tools | |
| session = agent.context_manager.session | |
| assert session is not None | |
| if __name__ == "__main__": | |
| pytest.main([__file__, "-v"]) | |