Dev
GitHub repos gaining traction - what high-signal users are starring and what's climbing the board, captured daily and enriched from GitHub. Raw material for spotting new tech and patterns worth building on.
1,506
repos tracked
261
surfaced this week
215
created < 30d
Python
top language
18 repos
-
Finetune Llama 3.1, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory
-
5X faster 60% less memory QLoRA finetuning
-
High-performance code intelligence MCP server. Indexes codebases into a persistent knowledge graph — average repo in milliseconds. 158 languages, sub-ms queries, 99% fewer tokens. Single static binary, zero dependencies.
-
Finetune Llama 3.1, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory
-
A high-throughput and memory-efficient inference and serving engine for LLMs
-
Finetune Llama 4, DeepSeek-R1, Gemma 3 & Reasoning LLMs 2x faster with 70% less memory! 🦥
-
A memory efficient string type that can store up to 24* bytes on the stack
-
Supervised OTP runtime for Alloy — sessions, async dispatch, memory stores
-
Stateful agents that are like people, with memory, identity, and the ability to learn and adapt
-
Agent memory for LLMs: 30 runnable Jupyter notebooks covering conversation buffers, vector stores, knowledge graphs, episodic and semantic memory, MemGPT, Mem0, Letta, Zep, Graphiti, LoCoMo benchmarks, and production patterns.
-
Open source code for ICLR 2026 Paper: Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions
-
Benchmarking Chat Assistants on Long-Term Interactive Memory (ICLR 2025)
-
Fast and memory-efficient exact attention
-
The end of web parsing. The beginning of scalable pixel-native search. link: https://pixelrag.ai/
-
A modern replacement for Redis and Memcached
-
A Mac-compatible AI partner with memory. — local computer access, browser control, IM channels, MCP-native. Built on Shannon.
-
Desktop version of Pytorch memory viz, handling profiles of 10GB and more, with quality-of-life improvements