lightweight and on-device AI runtimes
Recent window: last 5.2 months
(2026-04-24 → 2026-09-28), compared with the prior
5.2 months.
horizontal · 264 members ·
Data as of 2026-09-30
These products provide compact models, distillation techniques, and optimized runtime engines designed to run AI locally on consumer hardware and edge devices. They are built for developers, hardware hackers, and embedded systems engineers seeking private, low-latency machine learning execution without expensive cloud GPUs. Unlike general cloud-hosted LLM APIs, this cluster focuses strictly on extreme memory efficiency and resource-constrained local inference.
Metrics
- Stage
- crowded
- Recent count
- 134
- Prior count
- 111
- Total count
- 264
- Momentum
- 20.72
- Attention
- 0.54
- Crowding
- 0.70
- Concentration
- 0.83
- Opportunity
- 0.33
Opportunity components
- Attention
- 0.54
- Low crowding
- 0.30
- Momentum (normalized)
- 0.30
- Low concentration
- 0.17
Source split
- github
- 10 (0.07)
- hn
- 91 (0.68)
- ph
- 33 (0.25)
- yc
- 0 (0.00)
Dominant source: hn · Divergence: 0.68
Members
|
Name
|
Source
|
Upvotes ▲
|
Launched
|
| 500k+ events/sec transformations for ClickHouse ingestion |
HN |
13 |
2026-04-08 |
| Llama.cpp Tutorial 2026: Run GGUF Models Locally on CPU and GPU |
HN |
13 |
2026-04-18 |
| Token Economics Calculator for AI inference hardware |
HN |
13 |
2025-11-19 |
| gguf-head |
GITHUB |
14 |
2026-09-16 |
| ExANS |
HN |
15 |
2026-08-05 |
| I built Wool, a lightweight distributed Python runtime |
HN |
15 |
2026-03-14 |
| Mqtt Broker for 10 Years |
HN |
15 |
2026-06-01 |
| Local text, image, video, music and 3D from one CLI, no Python |
HN |
16 |
2026-07-30 |
| LocalLLM |
HN |
16 |
2026-04-23 |
| HoundDog.ai |
HN |
16 |
2026-02-02 |
| An LLM response cache that's aware of dynamic data |
HN |
17 |
2026-01-07 |
| I built an open-source Linux-capable single-board computer with DDR3 |
HN |
17 |
2025-12-24 |
| hk |
GITHUB |
17 |
2026-09-13 |
| Open-weight OCR got so cheap I had to share it |
HN |
17 |
2026-07-24 |
| I made an open-source Rust program for memory-efficient genomics |
HN |
17 |
2025-11-13 |
| Run 500B+ Parameter LLMs Locally on a Mac Mini |
HN |
17 |
2026-03-09 |
| Hibana |
HN |
18 |
2026-02-06 |
| I built a lite LPU that can do inference on Karpathy's MicroGPT |
HN |
18 |
2026-08-24 |
| Luxonis |
HN |
18 |
2025-12-11 |
| YourMemory, agentic memory is a pruning problem, not a hoarding problem |
HN |
19 |
2026-06-07 |
| Cj–tiny no-deps JIT in C for x86-64 and ARM64 |
HN |
21 |
2025-11-05 |
| Tokenflood |
HN |
21 |
2025-11-12 |
| Cuts Long Horizon Inference Costs by 50% via external KV Cache Offload |
HN |
22 |
2026-07-26 |
| Tensor Spy: inspect NumPy and PyTorch tensors in the browser, no upload |
HN |
22 |
2026-03-02 |
| Run GUIs as Scripts |
HN |
22 |
2026-04-10 |
| Shibuya |
HN |
22 |
2026-02-23 |
| Running PrismML's Bonsai inside DRAM by breaking DDR4 timing rules |
HN |
23 |
2026-07-23 |
| Bsub.io |
HN |
23 |
2025-11-17 |
| BVisor |
HN |
24 |
2026-02-23 |
| Watch 14-Byte AI "brains" attempt to solve a 2D maze (Its hard) |
HN |
25 |
2026-07-27 |
| I indexed 8,643 BSides talks across 227 chapters and 6 continents |
HN |
25 |
2026-05-04 |
| Optimizing LiteLLM with Rust |
HN |
27 |
2025-11-18 |
| Conway's Game of Life in boot sector |
HN |
27 |
2026-09-21 |
| SHADOW-50M-Instruct |
GITHUB |
29 |
2026-09-15 |
| fast-long-context |
GITHUB |
29 |
2026-09-20 |
| Swift-Qwen3.8-27B, -58.3% thinking, x1.95 speed, accuracy of xhigh |
HN |
30 |
2026-09-16 |
| High speed graphics rendering research with tinygrad/tinyJIT |
HN |
31 |
2026-01-22 |
| E80: an 8-bit CPU in structural VHDL |
HN |
34 |
2026-01-17 |
| I created a RAW to HDRI stacker in (mostly) Common Lisp |
HN |
35 |
2026-06-05 |
| Linggen |
HN |
36 |
2025-12-19 |
| What is HN thinking? Real-time sentiment and concept analysis |
HN |
37 |
2026-02-12 |
| Gerbil |
HN |
37 |
2025-11-11 |
| Low-latency local LLM runner via OpenJDK Panama FFM (Java 22) |
HN |
38 |
2026-07-14 |
| Python SDK |
HN |
43 |
2025-12-18 |
| I ran a language model on a PS2 |
HN |
46 |
2026-03-21 |
| Cancer diagnosis makes for an interesting RL environment for LLMs |
HN |
46 |
2025-11-12 |
| I built a RISC-V emulator that runs DOOM |
HN |
50 |
2026-05-03 |
| A pure ARM64 Assembly web server, now on Linux with CGI for no reason |
HN |
51 |
2026-06-23 |
| DOOM in the kernel, or fibers in eBPF |
HN |
53 |
2026-09-09 |
| NanoEuler |
HN |
55 |
2026-06-28 |