Which has better activity and maintenance signals?

Based on the RepoPilot analysis provided, mlc-ai/mlc-llm has stronger maintenance signals overall:

mlc-ai/mlc-llm:

• 23,081 stars, 2,115 forks — substantial community adoption
• Verdict: GO — Healthy across all four use cases — explicit positive assessment
• OpenSSF Scorecard: 4.3/10 (weak but present)
• Only 1 moderate CVE in direct dependencies (sentencepiece), 0 critical/high
• Active dependency scanning from deps.dev

significant-gravitas/autogpt:

• No quantitative maintenance signals provided in the repository data
• The CONTRIBUTING.md file is current and welcoming, suggesting active governance
• Analytics SQL query [autogpt_platform/analytics/queries/users_activities.sql:L1-L40] shows infrastructure for tracking user engagement
• Test files exist (e.g., helpers_test.py, dependencies_test.py), indicating ongoing testing practices

Key difference: mlc-ai/mlc-llm has explicit computed signals from RepoPilot's analysis (star count, fork count, dependency audit, security scorecard, and explicit "GO" verdict), whereas autogpt's signals are implicit in its codebase structure and contribution guidelines. mlc-llm's larger star/fork counts and explicit security scanning also suggest broader visibility and more active community engagement.

FOLLOW_UPS:

• What specific mlc-llm features or deployment targets matter for your use case?
• Does autogpt's agent graph execution architecture align with your needs?
• Which repo's dependency profile (ML compilation vs. backend platform) fits better?