Alternatives
Products that do what Serve 100 Large AI models on a single GPU with low impact to TTFT does
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…
- 1
General Compute▲315AI models that run on an inference cloud optimized for speed
May 2026 · generalcompute.com
- 2
- 3
- 4IP
The stack: two agents on separate boxes. The public one (nullclaw) is a 678 KB Zig binary using ~1 MB RAM, connected to an Ergo IRC server. Visitors talk to it via a gamja web client embedded in my site. The private one (ironclaw) handles email and scheduling, reachable only over Tailscale via Google's A2A protocol. Tiered inference: Haiku 4.5 for conversation (sub-second, cheap), Sonnet 4.6 for tool use (only when needed). Hard cap at $2/day. A2A passthrough: the private-side agent borrows the gateway's own inference pipeline, so there's one API key and one billing relationship…
Mar 2026 · georgelarson.me
- 5
- 6TV
May 2026 · github.com
- 75L
We've built InferX, a specialized runtime environment that fundamentally changes how LLMs are served. The core problem we solve is the latency bottleneck in AI inference, especially with large models. Current systems waste resources or suffer from painfully slow cold starts. InferX's AI-native architecture, with its "snapshot" technology, enables: * *Sub-2s cold starts:* Spin up models instantly. * *High density:* Serve more LLMs on the same GPUs. * *Optimal efficiency:* Maximize GPU utilization. This isn't just another API; it's a new execution layer designed from the ground up for the…
2025 · github.com
- 8
- 9AT
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.
27d ago · mikeayles.com
- 10IR
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
- 11
- 12TR
2018 · actcast.io
- 13GA
2021 · inferrd.com
- 14PA
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
- 15RA
Hi there, looking for feedback on my new project "Featherless.AI" The idea is to allow users to run all the models on hugging face instantly. Via the OpenAI API compatible endpoint. Why? Because its a real chore to download models and spin up GPUs, especially if you want to test multiple models. Not to mention GPUs cost multiple dollars an hour to rent. And if we want more people to use open source AI, we got to make it easier for them to try and play with all of them. So what if instead of spinning up dedicated GPUs per model (which is what every provider is doing) We can startup a LLM…
2024 · featherless.ai
- 16
High performance storage engine for efficient LLM inference and GPU Training.
17h ago · theopenlake.com
- 17CL
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
- 18IR
Jun 2026 · github.com
- 19RG
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
- 20MA
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:…
2023 · github.com
- 21
- 22AO
Hey hackers, the world needs more AI researchers with good taste, and hardcore software folks have some of the best. Many software friends mentioned they learn better from implementations than from papers, but existing open-source examples rarely go beyond basic nanoGPT-level demos. To help bridge that gap, I spent the last two months full-time reimplementing and open-sourcing a self-contained implementation of every major modern deep learning technique from scratch. The result is beyond-nanoGPT, containing 20k+ lines of handcrafted, minimal, and extensively annotated PyTorch code. I'd love…
2025 · github.com
- 23ML
Aug 2026 · github.com
- 24CM
Hey HN, I've been building AutoAgents, an AI agent framework in Rust. Today I'm sharing a feature I haven't seen done well elsewhere: composable middleware layers for LLM inference pipelines. The problem Every agent framework lets you swap LLM providers. Almost none of them give you a structured way to enforce safety, caching, or data sanitization in the inference path itself. You end up with guardrails as application-level if-statements, caching bolted on as a separate service, and PII handling as a "we'll add it later" TODO that never ships. This gets worse with local models. Cloud APIs…
Mar 2026 · github.com
Ranked by how close each launch is in meaning, then by votes. Refine with a description →