Alternatives
Products that do what TurboQuant-MoE:8.5x KV-Cache Compression does
8.5x KV-cache compression for LLM inference
- 1KR
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…
2025 · github.com
- 2SU
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
- 3
- 48F
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
- 5
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
- 6IV
The video demo runs a 7b Model on a normal gaming GPU. I think it already works quite well (accounting for the limited hardware power). :)
2024 · github.com
- 7TV
May 2026 · github.com
- 8
- 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.
28d ago · mikeayles.com
- 11
High performance storage engine for efficient LLM inference and GPU Training.
1d ago · theopenlake.com
- 12
- 13
- 14WM
Try it out! https://glhf.chat/ Hey HN! We’ve been working for the past few months on a website to let you easily run (almost) any open-source LLM on autoscaling GPU clusters. It’s free for now while we figure out how to price it, but we expect to be cheaper than most GPU offerings since we can run the models multi-tenant. Unlike Together AI, Fireworks, etc, we’ll run any model that the open-source vLLM project supports: we don’t have a hardcoded list. If you want a specific model or finetune, you don’t have to ask us for it: you can just paste the Hugging Face link in and…
2024 · glhf.chat
- 15
Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU. - MakazhanAlpamys/Soup
Aug 2026 · github.com
- 16
- 17
Powers faster, efficient reasoning for long-running agents
Jun 2026 · developer.nvidia.com
- 18
- 19
- 20OA
I built an experiment that uses an overfitted transformer and arithmetic coding to compress individual files. Instead of training the model to generalize, I train a 900KB transformer to memorize a single file and predict the next byte. Those predictions are fed into an arithmetic coder to produce the compressed output. On a 100MB NYC taxi CSV, it compresses to about 7MB (~0.5 bits/byte). On a 100MB slice of enwik9, it compresses to about 21MB (~1.68 bits/byte). It's pretty slow right now (roughly 20–30 minutes of training and 45 minutes each for compression and decompression on my…
Jun 2026
- 21
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
- 22EL
2023 · github.com
- 23KP
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
- 24AN
Kimi K3 has 2.78 trillion parameters and ships as 1.42 TB of weights. It clearly does not fit in the memory of a laptop. But K3 is a Mixture-of-Experts model. For each token, only a small fraction of its 896 experts per layer is activated. That changes the problem: the entire model does not need to be resident in RAM, as long as the weights required by each token can be reached quickly enough. We built WASTE — the Weight-Aware Streaming Tensor Engine — to explore that idea. WASTE keeps the dense, repeatedly used part of the model resident in memory, stores the routed experts in an…
Jul 2026
Ranked by how close each launch is in meaning, then by votes. Refine with a description →