AI · alternatives · 2026
24 alternatives to KVSplit – Run 2-3x longer contexts on Apple Silicon
I discovered that in LLM inference, keys and values in the KV cache have very different quantization sensitivities. Keys need higher precision than values to maintain quality. I patched llama.cpp to enable different bit-widths for keys vs. values on Apple Silicon. The results are surprising: - K8V4 (8-bit keys, 4-bit values): 59% memory reduction with only 0.86% perplexity loss - K4V8 (4-bit keys, 8-bit values): 59% memory reduction but 6.06% perplexity loss - The configurations use the same number of bits, but K8V4 is 7× better for quality This means you can run LLMs with 2-3× longer…
KVSplit is a modification to llama.cpp that enables independent quantization of keys and values in the KV cache for Apple Silicon processors. By using higher precision for keys and lower precision for values, it… Below are 24 products that do a similar job, ranked by how close each is in meaning and then by launch-day votes.
- 1
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 →
- 2
Hi HN! I'm just sharing a project I've been working on during the LLM Efficiency Challenge - you can now finetune Llama with QLoRA 5x faster than Huggingface's original implementation on your own local GPU. Some highlights: 1. Manual autograd engine - hand derived backprop steps. 2. QLoRA / LoRA 80% faster, 50% less memory. 3. All kernels written in OpenAI's Triton language. 4. 0% loss in accuracy - no approximation methods - all exact. 5. No change of hardware necessary. Supports NVIDIA GPUs since 2018+. CUDA 7.5+. 6. Flash Attention support via Xformers. 7. Supports 4bit and 16bit…
2023 · github.com · 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
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 →
- 5
Almost exactly 1 year ago, I submitted something to HN about using Llama2 (which had just come out) to improve the output of Tesseract OCR by correcting obvious OCR errors [0]. That was exciting at the time because OpenAI's API calls were still quite expensive for GPT4, and the cost of running it on a book-length PDF would just be prohibitive. In contrast, you could run Llama2 locally on a machine with just a CPU, and it would be extremely slow, but "free" if you had a spare machine lying around. Well, it's amazing how things have changed since then. Not only have models gotten a lot better,…
2024 · github.com · its alternatives →
- 6
I spent a lot of time and money on this rather big side project of mine that attempts to replicate the mechanistic interpretability research on proprietary LLMs that was quite popular this year and produced great research papers by Anthropic [1], OpenAI [2] and Deepmind [3]. I am quite proud of this project and since I consider myself the target audience for HackerNews did I think that maybe some of you would appreciate this open research replication as well. Happy to answer any questions or face any feedback. Cheers [1]…
2024 · github.com · its alternatives →
- 7
We've open-sourced Klarity - a tool for analyzing uncertainty and decision-making in LLM token generation. It provides structured insights into how models choose tokens and where they show uncertainty. What Klarity does: - Real-time analysis of model uncertainty during generation - Dual analysis combining log probabilities and semantic understanding - Structured JSON output with actionable insights - Fully self-hostable with customizable analysis models The tool works by analyzing each step of text generation and returns a structured JSON: - uncertainty_points: array of {step, entropy,…
2025 · github.com · its alternatives →
- 8TF
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 →
- 9
- 10AT
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 →
- 11
Working on Mac, Linux, and Windows now. I include a simple GUI to find new models and get things built and set up. It is working quite well across a few models for me. The GitHub README and DESIGN.md files go into detail of the how/why and it's working remarkably well so far. https://github.com/notactuallytreyanastasio/shoehorn
Aug 2026 · notactuallytreyanastasio.github.io · its alternatives →
- 12
Hey Folks! I've been building an open source benchmark for measuring local LLM performance on your own hardware. The benchmarking tool is a CLI written on top of Llamafile to allow for portability across different hardware setups and operating systems. The website is a database of results from the benchmark, allowing you to explore the performance of different models and hardware configurations. Please give it a try! Any feedback and contribution is much appreciated. I'd love for this to serve as a helpful resource for the local AI community. For more check out: - Website:…
2025 · localscore.ai · its alternatives →
- 13AT
I recently built a small open-source tool to benchmark different LLM API endpoints — including OpenAI, Claude, and self-hosted models (like llama.cpp). It runs a configurable number of test requests and reports two key metrics: • First-token latency (ms): How long it takes for the first token to appear • Output speed (tokens/sec): Overall output fluency Demo: https://llmapitest.com/ Code: https://github.com/qjr87/llm-api-test The goal is to provide a simple, visual, and reproducible way to evaluate performance across different LLM providers, including…
2025 · llmapitest.com · its alternatives →
- 14
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 →
- 15
G'day, HN! I'm one of the maintainers of `llm`. I've been working alongside a trusty group of contributors to bring this project to life, and we're now at a point where we're ready to share it with the world. Large language models (LLMs) are taking the computing world by storm due to their emergent abilities that allow them to perform a wide variety of tasks, including translation, summarization, code generation, and even some degree of reasoning. However, the ecosystem around LLMs is still in its infancy, and it can be difficult to get started with these models. `llm` is a one-stop shop for…
2023 · github.com · its alternatives →
- 16MG
Hello HN, I've been working on this project for a while, and it has been in an "open" beta for some time. I finally believe it's ready for its first release. I hope you like it. Here are some potential questions that may arise: 1. How does it compare to LM Studio? It's likely that if you're already using LM Studio, you'll continue to do so. This project is designed to be more user-friendly. 2. Is it open-source? No, it is not. 3. Does it use any open-source libraries? Yes, it uses llama.cpp and a few others, as indicated in the license information included with the application. 4. Why is not…
2023 · avapls.com · its alternatives →
- 17AB
we've been working on a KV store for the past year or so which is 2-6x faster than Redis (benchmark link below) yet disk persisted! so you get the speed of in-memory KV stores but with disk persistence. To achieve this we've created our custom filesystem that is optimized for our special usecase and we're doing smart batching for writes and predictive fetching for reads. In addition to basic operations, it also provides atomic inc/dec, atomic json patch, range scans and a unique key monitoring mechanism (pub-sub) over WebSockets which essentially allows you to receive notification on…
2025 · hpkv.io · its alternatives →
- 18
I work on inference scheduling — KV cache-aware routing, load balancing across GPU workers, that kind of thing. I wanted something like k9s but for my inference stack. Nothing existed, so I built it. llmtop is a real-time terminal dashboard for LLM inference workers. It scrapes the Prometheus /metrics endpoints that vLLM, SGLang, and LMCache already expose and shows everything in one view: KV cache usage, queue depth, TTFT/ITL latencies (P50/P99 from histogram buckets), token throughput, prefix cache hit rates. Color-coded — red means go fix it. ``` brew install…
Mar 2026 · github.com · its alternatives →
- 19
Hey HN, we’re the developers of OpenLake, an open source storage engine for offloading LLM KV caches from GPU memory into a shared tier of RAM and NVMe. We built OpenLake because KV caches are outgrowing GPU memory. A single 256K token conversation on Gemma 4 31B produces approximately 43GB of KV state, more than half the memory of an 80GB H100. The problem becomes even harder across a cluster: a prefix cached on one GPU host is unavailable when the next request lands on a different GPU, forcing the new GPU to repeat work the fleet has already completed. Once the KV cache is offloaded,…
Jul 2026 · github.com · its alternatives →
- 20WM
Pick a Mac and a model, get the memory it needs and the tokens/sec to expect. Covers M1 to M6, every unified memory size, Q4 to FP16. Formula fitted and validated on our own measurements on two machines.
6d ago · macyou.co · its alternatives →
- 21
Hey HN, we're super excited to share something we've been working on: OpenLIT. After an engaging preview that some of you might recall, we are now proudly announcing our first stable release! *What's OpenLIT?* Simply put, OpenLIT is an open-source tool designed to make monitoring your Large Language Model (LLM) applications straightforward. It’s built on OpenTelemetry, aiming to reduce the complexities that come with observing the behavior and usage of your LLM stack. *Beyond Basic Text Generation:* OpenLIT isn’t restricted to just text and chatbot outputs. It now includes automatic…
2024 · github.com · its alternatives →
- 22
Hey, recently I took inspiration from llama.cpp, ollama, and many other similar tools that enable inference of LLMs locally, and I just finished building a Llama inference engine for the 8B model in CUDA C. I recently wanted to explore my newly founded interest in CUDA programming and my passion for machine learning. This project only makes use of the native CUDA runtime api and cuda_fp16. The inference takes place in fp16, so it requires around 17-18GB of VRAM (~16GB for model params and some more for intermediary caches). It doesn’t use cuBLAS or any similar libraries since I wanted to be…
2025 · github.com · its alternatives →
- 23
I thought it'd be interesting to use Linux PSI (Pressure Stall Information) for an LLM runtime to trim the KV cache. This is mainly useful imo for edge devices like the Jetson Orin super nano kit which have unified memory. I haven't benched much, but plan to do so more over time and see if I can make a real use of it as I run local LLMs. Let me know if it makes sense :P (I of course vibed this idea)
Jun 2026 · github.com · its alternatives →
- 24
I wanted to run AI from inside the JVM. I started out with the standard REST sidecar, ripped that out to use Project Panama (Foreign Function & Memory API) in the new JDK versions to interface directly with llama.cpp. I still wasn't happy with how that functioned, so I built libargus.cc to get a clean ABI to expose a structured API up in the JVM landscape. It still uses Project Panama to interface directly with llama.cpp, whisper.cpp, and ggml compute graphs. I have zero-allocation on the hot paths, memory segments for prompts and tokens are allocated once inside confined Arenas. Raw…
Jul 2026 · github.com · its alternatives →
Also compare
- Speeding up LLM inference 2x times (possibly) alternatives
- 80% faster, 50% less memory, 0% loss of accuracy Llama finetuning alternatives
- Tiny-vLLM – high performance LLM inference engine in C++ and CUDA alternatives
- We made our own inference engine for Apple Silicon alternatives
- LLM-aided OCR – Correcting Tesseract OCR errors with LLMs alternatives
- Llama 3.2 Interpretability with Sparse Autoencoders 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 →