BitLlama

imonoonoko · imonoonoko.BitLlama

Pure Rust LLM inference engine with 1.58-bit ternary support and Test-Time Training

BitLlama is a Pure Rust LLM inference engine featuring 1.58-bit ternary quantization, Test-Time Training (TTT), Soul learning system, MCP server/client, and private RAG. Supports Llama, Gemma, Mistral, Qwen, and BitNet models. OpenAI-compatible API server included.

winget install --id imonoonoko.BitLlama --exact --source winget

Latest 1.0.0

Release Notes

v1.0.0 — Final Release BitLlama v1.0.0. Development complete. What is BitLlama? A Pure Rust LLM inference engine with Soul learning and hierarchical memory.

  • 7 model architectures: Llama-2/3, Gemma-2/3, Qwen2.5, Mistral, BitNet
  • Soul learning: LoRA fine-tuning from conversations
  • Memory system: 4-layer hierarchical memory + 7-stage Sleep consolidation
  • Desktop GUI: Tauri 2.0 + Svelte 5, Japanese/English i18n
  • Performance: 45.4 tok/s (7B), 90% of llama.cpp
  • 1121 tests, quality score 9.0/10 Changes since v0.16.0
  • CJK memory search fix (character bigram fallback for Japanese queries)
  • Soul learning tests (warmup, chat template, VRAM guard)
  • Chat template application fix for GGUF tokenizer fallback
  • README/ROADMAP updated to reflect project completion Install

Homebrew

brew tap imonoonoko/bitllama && brew install bitllama

winget

winget install imonoonoko.BitLlama

Or download binaries below

Built with Rust by @imonoonoko Full Changelog: v0.16.0...v1.0.0

Installer type: portable

Architecture Scope Download SHA256
x64 Download 4F2A1FC7F498F43292E52F32DCF4E88B5AD4C0DFED933AAA5815102CF8D6DCA7

Details

Homepage
https://github.com/imonoonoko/Bit-TTT-Engine
License
MIT
Publisher
imonoonoko
Support
https://github.com/imonoonoko/Bit-TTT-Engine/issues

Tags

aicliinferencellmmachine-learningrust

Older versions (2)

0.16.0
Architecture Scope Download SHA256
x64 Download C8E7AAF01655D0342241C613181EFAF9A430EC19FC558BE9CA639394A3D9BE27
0.15.0
Architecture Scope Download SHA256
x64 Download 5652106ED45DB2B8963CACA58739BAF8F2C5DCD4F034C6CD4A4F33FAC5C72A3B