Life & fun · alternatives · 2026
24 alternatives to Moondream, a small vision language model that runs on 8GB of RAM
I've been working on training this small vision language model for the last month - excited to release the first prototype today! It is based on SigLIP (image encoder), Phi-1.5 (text model) and trained using the LLaVa-1.5 training dataset. It runs reasonably fast on CPU with ~8GB of RAM in full 32-bit precision. There's plenty of room to speed it up and reduce memory consumption by quantizing the model. I posted a video of it running on my M2 Macbook Air (on CPU not MPS, so performance should be comparable on other hardware) on Twitter to demonstrate inference speed:…
Below are 24 products that do a similar job, ranked by how close each is in meaning and then by launch-day votes. Moondream, a small vision language model that runs on 8GB of RAM launched in 2023; newer entries below may have overtaken it.
- 1
Built a ~9M param LLM from scratch to understand how they actually work. Vanilla transformer, 60K synthetic conversations, ~130 lines of PyTorch. Trains in 5 min on a free Colab T4. The fish thinks the meaning of life is food. Fork it and swap the personality for your own character.
Apr 2026 · github.com · its alternatives →
- 2
Here's a project I've been working on for the last few months. It's a new (I think) algorithm, that allows to adjust smoothly - and in real time - how many calculations you'd like to do during inference of an LLM model. It seems that it's possible to do just 20-25% of weight multiplications instead of all of them, and still get good inference results. I implemented it to run on M1/M2/M3 GPU. The mmul approximation itself can be pushed to run 2x fast before the quality of output collapses. The inference speed is just a bit faster than Llama.cpp's, because the rest of implementation…
2024 · asciinema.org · its alternatives →
- 3
Build your own high performance LLM inference engine in C++ and CUDA - a smaller version of vLLM - jmaczan/tiny-vllm
May 2026 · github.com · its alternatives →
- 4
I wanted to share our new speech to text model, and the library to use them effectively. We're a small startup (six people, sub-$100k monthly GPU budget) so I'm proud of the work the team has done to create streaming STT models with lower word-error rates than OpenAI's largest Whisper model. Admittedly Large v3 is a couple of years old, but we're near the top the HF OpenASR leaderboard, even up against Nvidia's Parakeet family. Anyway, I'd love to get feedback on the models and software, and hear about what people might build with it.
Feb 2026 · github.com · its alternatives →
- 5
DeepSeek-V4▲387The open-source era of 1M context intelligence
Apr 2026 · huggingface.co · its alternatives →
- 6IB
We wanted to do something very challenging to prove to ourselves that we can do anything we put our mind to. The reasoning for why we chose to build a toy TPU specifically is fairly simple: - Building a chip for ML workloads seemed cool - There was no well-documented open source repo for an ML accelerator that performed both inference and training None of us have real professional experience in hardware design, which, in a way, made the TPU even more appealing since we weren't able to estimate exactly how difficult it would be. As we worked on the initial stages of this project, we…
2025 · tinytpu.com · its alternatives →
- 7AT
A 3.16M-parameter INT4 transformer running entirely in the on-chip memory of a Xilinx Kria KV260. Zero DRAM in the token loop, 59,965 tok/s on the fabric, bit-exact. Chat with it live.
Aug 2026 · mikeayles.com · its alternatives →
- 8
We wrote our inference engine on Rust, it is faster than llama cpp in all of the use cases. Your feedback is very welcomed. Written from scratch with idea that you can add support of any kernel and platform.
2025 · github.com · its alternatives →
- 9
A simple mobile web app inspired by Fuzzy-Search/realtime-bakllava that uses llama.cpp server backend with multimodal mode to describe and narrate what the phone camera sees. I built this thing in a few hours using a single ChatGPT thread to generate most things for me and iterate on this project. Here's the workflow: https://chat.openai.com/share/ea84ec69-5617-45e8-8772-ac2dcf...
2023 · github.com · its alternatives →
- 10
Long-context efficiency with DeepSeek Sparse Attention
2025 · huggingface.co · its alternatives →
- 11
The vision plugin for OpenCode that truly understands images. Inspect, read, and reason about any screenshot or picture with deeper understanding than any other plugin — fully local, private, and free. Normally, it takes 300ms to analyse one image on my laptop, which is pretty fast for a local vision model. I use moondream2 as my vision model, you can set your custom model like moondream3.1 if you have a good GPU (for comparison I have currently have an RTX 3050). It works cross-platform. Just follow the README. If you like my work, you leave me a tip as an act for supporting open source!!…
Aug 2026 · github.com · its alternatives →
- 12
The Emotion Engine has 32 MB of RAM total, so the trick is streaming weights from CD-ROM one matrix at a time during the forward pass — only activations, KV cache and embeddings live in RAM. This means models bigger than the RAM can still run, they just read more from disc. Had to build a custom quantized format (PSNT), hack endianness, write a tokenizer pipeline, and most of the PS2 SDK from scratch (releasing that separately). The model itself is also custom — a 10M param Llama-style architecture I trained specifically for this. And it works. On real hardware.
Mar 2026 · github.com · its alternatives →
- 13
Minimal, readable LLM post-training experiments on one 8GB GPU. Measures forgetting, seed variance, and RL emergence. - pochenai/nano-llm-posttraining
Aug 2026 · github.com · its alternatives →
- 14
Most inference UIs that I've come across pretty much just give us a chat-like interface to toy around with models in a single visual conversation thread. Given the fact that we are limited to seeing only one output at a time, it's kind of hard to compare outputs from different models, adjustments made to the prompting, and sampler settings. But even when keeping the generation parameters the same (e.g., to test for reliability in the output) and just going for multiple passes, there is no easy way to have a side-by-side comparison to keep track of the outputs from the multiple "rounds". I…
2024 · github.com · its alternatives →
- 15
I wanted to know how fast a 26B mixture-of-experts model could run on a desktop CPU with no GPU. Got ~40 tok/s single-stream (lossless) and ~124 batched. The surprising part was the byte budget: for this model you compress the output head (32% of per-token bytes), not the experts (16%). The writeup has the bandwidth roofline and the dead-ends; the repo has the reproducible recipe. Happy to answer questions. Repo: https://github.com/arun-prasath2005/gemma4-cpu-moe
Jun 2026 · apeg.dev · its alternatives →
- 16
- 17
I wanted to build an inference provider for proprietary AI models, but I did not have a huge GPU farm. I started experimenting with Serverless AI inference, but found out that coldstarts were huge. I went deep into the research and put together an engine that loads large models from SSD to VRAM up to ten times faster than alternatives. It works with vLLM, and transformers, and more coming soon. With this project you can hot-swap entire large models (32B) on demand. Its great for: Serverless AI Inference Robotics On Prem deployments Local Agents And Its open source. Let me know if anyone…
Nov 2025 · github.com · its alternatives →
- 18
It's our new text-to-image model: a 9.3B single-stream diffusion transformer trained entirely from scratch. We focused heavily on controllability through structured JSON prompts, with strong text rendering, spatial awareness through bounding box guidance, and color palette control. It has the best text rendering of any open-weight model we've tested so far, and the NF4 quantized checkpoint runs on a single 24GB GPU. For more technical details and examples see our blog post: https://ideogram.ai/blog/ideogram-4.0/ We will be happy to answer any questions :)
Jun 2026 · github.com · its alternatives →
- 19HI
元石科技官网 - 问小白 - 全能AI：搜得准 聊得嗨 读得懂，支持DeepSeek R1/V3/V3.1等最新大模型，支持 AI 图片生成，AI 图片编辑，AI 联网搜索。
2013 · yosee.com · its alternatives →
- 20TF
I’d originally launched my app: Private LLM[1][2] on HN around 10 months ago, with a single RedPajama Chat 3B model. The app has come a long way since then. About a month ago, I added support for 4-bit OmniQuant quantized Mixtral 8x7B Instruct model, and it seems to outperform Q4 models at inference speed and Q8 models at text generation quality, while consuming only about 24GB of RAM[3] at 8k context length. The trick is: a) to use a better quantization algorithm and b) to use unquantized embeddings and the MoE gates (the overhead is quite small). Other notable features include many more…
2024 · its alternatives →
- 21
Hi HN, I built OpenGraviton, an open-source AI inference engine that pushes the limits of running extremely large LLMs on consumer hardware. By combining 1.58-bit ternary quantization, dynamic sparsity with Top-K pruning and MoE routing, and mmap-based layer streaming, OpenGraviton can run models far larger than your system RAM—even on a Mac Mini. Early benchmarks: TinyLlama-1.1B drops from ~2GB (FP16) to ~0.24GB with ternary quantization. At 140B scale, models that normally require ~280GB fit within ~35GB packed. Optimized for Apple Silicon with Metal + C++ tensor unpacking, plus…
Mar 2026 · github.com · its alternatives →
- 22W1
Hey HackerNews, I built this project over the last few weeks as a palette cleanser from a failed game launch. I wanted to learn a bit about AI/Neural-Networks and naively thought I could build a tiny maze-solving AI in a weekend with a 100% solve rate. Well - I couldn't, but I got pretty close. 14 Bytes total model size, and a 96.5% solve rate on unseen mazes. Trained across 46 phases experimenting with different ideas to improve the model (better performance, smaller size). Its quite fun to watch the model attempt to solve the maze, when they fail its usually due to getting stuck in a…
Jul 2026 · con-dog.github.io · its alternatives →
- 23
Alt▲21Most AI notetakers have a 300 minute limit. This is not enough for uni students like me, or people who do remote work. So, I made a notetaker that runs the ASR model on-device. - Free with local ASR & LLM models - No transcription time limit - High accuracy for non-english speech, support for 100 languages - Real-time transcription - Zoom/Google Meet support - No internet needed - Efficient battery usage (~6 hours on full charge on my M2 Pro) This started as a small cli tool I made for myself, then many of my uni friends really liked it, so here it is as an app! I hope a lot of people…
Dec 2025 · altalt.io · its alternatives →
- 24PA
Hello Hacker News! I am Bertrand from Pruna AI. With my associates, John, Rayan, and Stephan, we are fellow researchers in AI efficiency and reliability coming from TUM. We are building an optimization engine that combines compression methods (e.g. quantization, pruning, compilation, batching…) in the aim of saving compute power when running AI models. This optimization engine take one base model as input and returns a compressed model as output. It aims to help for two things: - Make various AI models faster and/or smaller for various hardware (because they can require significant…
2024 · its alternatives →
Also compare
- I built a tiny LLM to demystify how language models work alternatives
- Speeding up LLM inference 2x times (possibly) alternatives
- Tiny-vLLM – high performance LLM inference engine in C++ and CUDA alternatives
- Moonshine Open-Weights STT models – higher accuracy than WhisperLargev3 alternatives
- DeepSeek-V4 alternatives
- I built a toy TPU that can do inference and training on the XOR problem alternatives
Ranked by how close each launch is in meaning, then by votes. Prices were read from each product’s own site when checked and can change. Refine with your own description →