← all editions
EDITION 84 · COLIBRI RUNS FRONTIER MOE MODELS IN PURE C AT 110 STARS/DAY2026·09·115 min readlinks verified live

Colibri runs frontier MoE models in pure C at 110 stars/day

We don't tell you what's popular — popularity lags and is gameable. We tell you what's gaining speed right now, and whether it's worth your attention. 6 accelerating AI repos earned today's slot.

↑110/day
fastest climber
in the edition
6
picks that
earned a slot
live
counts pulled
at publish
5min
to read the
whole edition
01

Top mover

★ TOP MOVER
JustVugg/colibriWATCHC▲ 110 /day★ 27,546

Colibri is a dependency-free C engine that runs large Mixture-of-Experts models by keeping only the router and shared weights in RAM while streaming expert weights from disk on demand, letting a laptop or single workstation serve models far larger than its memory. It's gaining traction because MoE architectures (Mixtral, DeepSeek, Qwen-MoE) make sparse activation practical, and disk-streaming sidesteps the VRAM wall that blocks local inference. The honest caveat: per-token latency is bound by storage throughput, so it trades speed for capacity and is unsuitable for high-QPS serving.

2mo oldApache-2.03k forksactive 1d ago116 open issues
Competes withllama.cpp
Who needs itDevelopers running large MoE models locally on memory-constrained hardware
02

🤖 Agents & automation

wang2122/sprix-sage-routerWATCHPython▲ 44 /day★ 3,668

Sprix Sage Router implements a state machine that classifies each incoming task as SELF, COLLABORATE, or HANDOFF, letting a network of agents decide locally whether to solve, co-solve, or delegate rather than hardcoding a static topology. It matters now because A2A-style inter-agent protocols are proliferating and most frameworks still route via fixed graphs or LLM-judge prompts, so a lightweight state-aware router is a plausible primitive. Caveat: the repo is young and the 3.6k stars look disproportionate to its maturity, so verify the routing policy, benchmarks, and failure modes before trusting it in production.

24d oldMIT422 forksactive 14d ago1 open issues
Competes withLangGraph
Who needs itEngineers building multi-agent orchestration layers who need dynamic task routing
tsingyuai/growth-labWATCHPython▲ 42 /day★ 1,999

growth-lab is an end-to-end agent pipeline that reads your product, pulls data from channels, researches the market, runs campaigns, and feeds results back into the next cycle. It ships as Claude Code / Codex skills, so it slots into existing agent workflows rather than being a standalone SaaS, and its Chinese-market focus (Xiaohongshu, SEO) is a real differentiator. The caveat: 'self-improving growth' depends heavily on the quality of your data connectors and eval loop, which the repo leaves largely to the user.

2mo oldApache-2.0169 forksactive 1mo ago0 open issues
Who needs itSolo founders and small growth teams running agent-driven marketing
03

Tooling & infra

plannotator/effective-htmlWATCHHTML▲ 73 /day★ 2,892

A collection of Claude Code / Codex agent skills that steer LLMs to emit self-contained HTML for wireframes, interactive prototypes, plans, and diagrams instead of Markdown or ASCII. It's gaining traction because HTML artifacts render natively in browsers, support real interactivity and layout, and sidestep the friction of image-based mockups or Mermaid limits. The caveat: it's a prompt/skill pack, not a runtime or library, so output quality still depends heavily on the underlying model and your ability to review generated markup.

3mo oldMIT200 forksactive 1mo ago3 open issues
Competes withMermaid, Excalidraw, Figma
Who needs itDevelopers and founders using Claude Code or Codex who want faster UI mockups and shareable plan docs
Albert-Weasker/niubigeoWATCHTypeScript▲ 66 /day★ 2,316

Niubigeo runs prompts against LLMs to measure how often a brand appears in AI-generated answers, then produces competitor visibility reports and GEO (generative engine optimization) recommendations. It's gaining traction because 'am I visible in ChatGPT/Perplexity answers' is a real new marketing channel with no standard measurement tooling yet. Caveat: the core method is prompt sampling, so results are noisy, non-deterministic, and easy to game — treat scores as directional, not ground truth.

7d oldApache-2.089 forksactive 2d ago11 open issues
Competes withProfound, Peec AI, Otterly.ai
Who needs itGrowth/marketing leads and SEO teams tracking brand presence in LLM answers
miqdadbadjuber/anti-slopWATCHJavaScript▲ 62 /day★ 2,096

A curated set of system-prompt rules that instruct coding agents (Cursor, Claude Code, etc.) to avoid telltale AI design patterns: purple gradients, generic hero sections, em-dash-heavy copy, and boilerplate code. It's gaining traction because the bottleneck has shifted from generation to taste — developers are drowning in plausible-but-soulless output and want a cheap, drop-in constraint layer. The caveat: it's a prompt/rules file, not a tool, so effectiveness depends entirely on your agent honoring instructions and your own ability to define what 'not slop' means.

1mo oldMIT145 forksactive 3d ago0 open issues
Who needs itDevelopers using AI coding agents who care about output taste

Catch the next breakout before it trends.

The fastest-accelerating open-source AI, curated and called. A fresh read every day. Free.

Join 8,400+ engineers · free · no spam
Almost there — check your inbox and click the confirmation link to finish.