Skip to content

Releases: ohnodev/obelisk-core

v0.2.0-alpha

05 Mar 09:59
@ohnodev ohnodev
v0.2.0-alpha
0900013
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Choose a tag to compare

View all tags
v0.2.0-alpha Pre-release
Pre-release

Alpha release (second alpha)

This release reflects the current Alpha status of Obelisk Core with updated docs, services overview, and Docker/PM2 guidance.

Highlights

  • README: Alpha status (v0.2.0-alpha), services diagram (Inference, Blockchain, Polymarket, Deployment API), Router URL and Inference Config example, Deployment API self-hosting note
  • Quick Start: Optional Blockchain/Polymarket startup section; Python 3.10+ and GPU clarified (only for local Qwen; Router users can skip)
  • Docker: INFERENCE_SERVICE_URL, BLOCKCHAIN_SERVICE_URL, POLYMARKET_SERVICE_URL in example; --add-host=host.docker.internal:host-gateway for Linux; Compose extra_hosts note
  • Templates: Sora template removed from UI (Aria kept); Clawballs description updated

Full changelog

See CHANGELOG.md for 0.2.0-beta and earlier entries.

Loading

v0.1.0-alpha - Initial Release

28 Jan 16:51
@ohnodev ohnodev
v0.1.0-alpha
300c91a

Choose a tag to compare

View all tags
v0.1.0-alpha - Initial Release Pre-release
Pre-release

[0.1.0-alpha] - 2025-01-28

Added

  • Self-hosted LLM: Qwen3-0.6B model with thinking mode support
  • Memory Layer: LangChain-based conversation memory with automatic summarization
  • Dual Storage Modes:
    • Solo mode (local JSON storage)
    • Prod mode (Supabase integration)
  • REST API: FastAPI server with endpoints for generation, memory, and evolution
  • CLI Interface: Interactive chat and command-line tools
  • Thinking Mode: Always-enabled thinking mode for enhanced reasoning (Qwen3)
  • Memory Management:
    • Buffer window memory for recent conversations
    • Automatic summarization every 10 message pairs
    • Context-aware responses using conversation history
  • Tests: Basic test suite with model output verification
  • Documentation:
    • README with installation and usage guides
    • CONTRIBUTING.md with PR guidelines
    • SECURITY.md with best practices
    • API.md, CLI.md, QUICKSTART.md
    • Test documentation

Technical Details

  • Python 3.8+ support
  • PyTorch-based inference with 4-bit quantization
  • LangChain integration for memory management
  • FastAPI for REST API
  • Click for CLI interface

Known Limitations

  • Alpha version - API may change
  • Evolution mechanics not yet fully tested
  • LoRA fine-tuning integration in progress
  • Quantum influence module planned
Loading