macaron-artifacts
Token-efficient tsx-based GenUI for Claude Code, Codex, OpenCode, OpenClaw, Hermes.
- 325
- TypeScript
- Apache-2.0
- 5 wk ago
MindLab-Research
build AI systems that learn from real experience — agents that remember, adapt, and act on behalf of the people who use them.
Built from MindLab-Research's public GitHub profile. Not on Take My Repo yet, so there are no posts or updates here.
Token-efficient tsx-based GenUI for Claude Code, Codex, OpenCode, OpenClaw, Hermes.
Generative UI for DeepSeek Harness — the agent writes TSX and the web UI compiles every frame as it streams, so a card renders while its JSX is still unclosed. Inline in the conversation, or in a side canvas.
A novel serving architecture for LoRA models — route, compose, and serve hundreds of LoRA adapters on a single base model.
Python toolkit for MinT, the open infrastructure for experiential intelligence and LoRA RL.
Repo of Paper: delta-Mem: Efficient Online Memory for Large Language Models
Open source version of Mind Lab's internal Megatron-LM fork for supporting GLM5.2
Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.
飞书 CLI 离线世界:373 typed methods, 486 shortcuts, 72 state collections, 668 verified tasks — Agent 执行环境的必备离线飞书 CLI 契约
Minimal multi-turn harness for delayed Sigma-memory reads after context compaction
ferrite — native GLM-5.3-Flash inference engine in Rust: PDAF disaggregation, composable TP/CP/DCP axis algebra, exact MHC, WYF chunkwise GatedDeltaNet, sm_100a CUDA kernels. CPU golden-standard verified (74 tests).