pagareajinkya04/llama-3.2-3b-instruct

fix: change FROM python:3.10-slim to 3.11-slim to force Docker cache invalidation

ci: trigger rebuild 1780921545

fix: default port to 7860 for HF Spaces compatibility

ci: force fresh Docker build 1780919939

ci: force rebuild 1780919584

fix: remove name-mangled __init_stabilization call from CircuitBreaker

fix: add missing get_planner, get_model_registry, SelfHealingRuntime imports

fix: resolve SyntaxError in logger string - remove stray quote from corrupted Unicode replacement

fix: remove invalid unicode characters from docstrings/comments

fix: add PORT env var fallback to FASTAPI_PORT in config

fix: remove --jinja flag, increase timeout to 600s, add system deps, improve startup handling

fix: remove stray 'n' causing NameError on module import

fix: add missing get_execution_graph() function to prevent ImportError on HF Spaces

fix: correct setup.py model download URL to Llama-3.2-3B-Instruct, remove duplicate FASTAPI_PORT in Dockerfile

fix: restrict CORS origins to env var instead of wildcard

fix: remove hardcoded JWT_SECRET fallback (change-me-in-production)

fix: Python 3.9 compatibility - Optional type hints and agent registration newlines (3 fixes: literal

fix: response.iter_bytes() -> response.aiter_bytes() for SSE streaming in proxy_openai

v1.0.2: Fixed Dockerfiles (libgomp1, removed chown timeout, selective COPY), 30s max timeout, HF_TOKEN_04 token, all 24 spaces LIVE/BUILDING verified

fix: Python 3.9 compatibility - Optional type hints and agent registration newlines (3 fixes: literal

Merge branch 'main' of https://huggingface.co/spaces/pagareajinkya04/llama-3.2-3b-instruct

fix: Python 3.9 compatibility - Optional type hints and agent registration newlines (3 fixes: literal

fix: python3.10, disable flash-attn, reduce memory for HF Spaces CPU

v1.0.2: Update llama.cpp b4600→b9279, enable FLASH_ATTN=True

Auto-update: sync enhanced model registry

fix: ADAM v4.5 - fix import indentation

fix: ADAM v4.5 - add AutonomousExecutionEngine import

fix: ADAM v4.5 runtime integration - imports, init, benchmarks

feat: ADAM v4.5 - unified autonomous API routes (/api/adam/autonomous/execute)

feat: ADAM v4.5 - evaluation, workflows, sandbox, reliability, learning, memory evolution, autonomous engine

feat: ADAM v4.5 - Multi-Agent Collaboration + Context Engineering

ADAM v4.0: autonomous execution intelligence platform

fix: await governance.get_status() in get_full_status - fixes HTTP 500

fix: correct get_health indentation - was nested local fn instead of class method

fix: await orchestrator.get_status() in get_full_status - fixes HTTP 500

perf: ADAM v3.1 stabilization - fix HTTP 500, heartbeat, HF sleep, peer scoring

fix: address code review - deps, auth wiring, heartbeat, status, middleware

feat: ADAM v3.0 lightweight production runtime - all 7 phases

fix: add missing await on node_intelligence.get_status() in production.py

fix: add missing await on governance.get_status() and orchestrator.get_status()

fix: correct constructor arg names in register_enhanced_routes

feat: production-hardened ADAM v3.0 Enhanced Runtime

fix: correct all truncation bugs in enhanced modules + remove masking try/except

fix: correct import path for agent_integration_enhanced module

feat: integrate ADAM v3.0 Enhanced Node Intelligence Runtime

fix: Python 3.9 compatibility - Optional type hints and agent registration newlines (3 fixes: literal

fix: IndentationError in app/main.py line 82 - literal backslash-n in agent registration block replaced with real newlines (caused Uvicorn crash, all HF Spaces down)

fix: SyntaxError in app/main.py line 14 - literal backslash-n replaced with real newline (caused Uvicorn crash, 503 on HF Spaces)

fix: Dockerfile invalid CMD instruction - commented out broken 'CMD to start' line (caused build failure on HF Spaces)

ADAM v2.0: Advanced Agentic Mesh — DAG orchestrator, cognition, knowledge web, forge tools, runtime optimization