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
|
| LyneCode Beats AntiGravity and Codex (and It's Open Source) |
HN |
7 |
2025-12-16 |
| Docker Model Runner Integrates vLLM for High-Throughput Inference |
HN |
7 |
2025-11-20 |
| We built a <60ms, open-source alternative to E2B using RustVMM and KVM |
HN |
7 |
2026-04-22 |
| Stillwind |
HN |
7 |
2026-06-11 |
| RiceVM |
HN |
7 |
2026-04-02 |
| I've hooked up 2D LiDARs to Raspberry Pi, wrote Python library lds2d |
HN |
7 |
2026-06-04 |
| I embedded 685M public texts in 32 minutes (on 8x A100, Rust, TensorRT) |
HN |
7 |
2026-06-04 |
| A new engine to run Kimi K3 on a laptop |
HN |
7 |
2026-07-29 |
| Composable middleware for LLM inference Optimization Passes |
HN |
7 |
2026-03-04 |
| RISCY-V02: A 16-bit 2-cycle RISC-V-ish CPU in the 6502 footprint |
HN |
7 |
2026-03-06 |
| A Wasm to Go Translator |
HN |
7 |
2026-02-26 |
| Linear RNN/Reservoir hybrid generative model, one C file (no deps.) |
HN |
7 |
2026-04-09 |
| Selora |
HN |
7 |
2026-06-17 |
| Memory Graph |
HN |
7 |
2026-01-05 |
| Veil |
HN |
7 |
2026-03-30 |
| Tsplat |
HN |
7 |
2026-04-14 |
| Dbzero |
HN |
7 |
2025-12-20 |
| agent-gpu-calculator |
GITHUB |
7 |
2026-09-24 |
| Unified multimodal memory framework, without embeddings |
HN |
7 |
2026-01-07 |
| EdgeVec |
HN |
7 |
2025-12-12 |
| Micro-RLE ≤264-byte compression for UART/MCU logs, zero RAM growth |
HN |
7 |
2025-11-01 |
| Serve 100 Large AI models on a single GPU with low impact to TTFT |
HN |
7 |
2025-11-08 |
| vw-firmware-mac |
GITHUB |
7 |
2026-09-12 |
| CJIT, a single-binary C compiler that can self host |
HN |
7 |
2026-04-13 |
| Python running on the Super Nintendo (in-browser demo) |
HN |
7 |
2026-07-06 |
| Trunchbull, run real models against any benchmark in your browser |
HN |
6 |
2026-08-12 |
| PrivateClaw |
HN |
6 |
2026-04-24 |
| Witchcraft and Pickbrain |
HN |
6 |
2026-04-16 |
| Flint |
HN |
6 |
2026-04-16 |
| OS Megakernel that match M5 Max Tok/w at 2x the Throughput on RTX 3090 |
HN |
6 |
2026-04-08 |
| Llama 3.2 3B and Keiro Research achieves 85% on SimpleQA |
HN |
6 |
2026-03-07 |
| Llmpm |
HN |
6 |
2026-03-09 |
| ChonkLM |
HN |
6 |
2026-05-09 |
| CodeDiff |
HN |
6 |
2026-09-29 |
| LLM Memory Storage that scales, easily integrates, and is smart |
HN |
6 |
2026-03-16 |
| NanoEuler |
HN |
6 |
2026-06-19 |
| A beautiful and local-first PDF reader for studying dense things |
HN |
6 |
2026-06-06 |
| Multi-agent autoresearch for ANE inference beats Apple's CoreML by 6× |
HN |
6 |
2026-03-31 |
| Micron: a high performance C++23 (re)implementation of Libc and the STL |
HN |
6 |
2026-06-05 |
| I generated a "stress test" of 200 rare defects from 7 real photos |
HN |
6 |
2026-02-12 |
| OpenEntropy |
HN |
6 |
2026-02-17 |
| Vicinae |
HN |
6 |
2026-07-08 |
| Zroar |
HN |
6 |
2026-08-21 |
| Shoehorn, a library to quantize an LLM to fit your Mac's VRAM |
HN |
6 |
2026-08-14 |
| Alcatraz |
HN |
6 |
2026-08-04 |
| Codex builds a working NES Emulator in one hour |
HN |
6 |
2026-02-26 |
| Semantic Overlays |
HN |
6 |
2026-09-01 |
| Blink-Edit |
HN |
6 |
2026-01-27 |
| Build apps with 500 models locally. No tracking, no cloud, just code |
HN |
6 |
2025-12-21 |
| Inference API that adapts to your SLA and quality constraints |
HN |
6 |
2026-01-02 |