Tune LLaMa3.1 on Google Cloud TPUs
Hey HN, we wanted to share our repo where we fine-tuned Llama 3.1 on Google TPUs. We’re building AI infra to fine-tune and serve LLMs on non-NVIDIA GPUs (TPUs, Trainium, AMD GPUs). The problem: Right now, 90% of LLM workloads run on NVIDIA GPUs, but there are equally powerful and more cost-effective alternatives out there. For example, training and serving Llama 3.1 on Google TPUs is about 30% cheaper than NVIDIA GPUs. But developer tooling for non-NVIDIA chipsets is lacking. We felt this pain ourselves. We initially tried using PyTorch XLA to train Llama 3.1 on TPUs, but it was rough: xla…
In plain words
This project provides tools and code for fine-tuning and serving Llama 3.1 on Google Cloud TPUs as an alternative to NVIDIA GPUs. It targets machine learning developers and organizations seeking cost-effective infrastructure for large language model workloads. The work addresses a gap in developer tooling for non-NVIDIA hardware by demonstrating how to run LLM training on TPUs, reportedly at approximately 30% lower cost than traditional GPU-based approaches.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hey HN, we wanted to share our repo where we fine-tuned Llama 3.1 on Google TPUs. We’re building AI infra to fine-tune and serve LLMs on non-NVIDIA GPUs (TPUs, Trainium, AMD GPUs). The problem: Right now, 90% of LLM workloads run on NVIDIA GPUs, but there are equally powerful and more cost-effective alternatives out there. For example, training and serving Llama 3.1 on Google TPUs is about 30% cheaper than NVIDIA GPUs. But developer tooling for non-NVIDIA chipsets is lacking. We felt this pain ourselves. We initially tried using PyTorch XLA to train Llama 3.1 on TPUs, but it was rough: xla integration with pytorch is clunky, missing libraries (bitsandbytes didn't work), and cryptic HuggingFace errors. We then took a different route and translated Llama 3.1 from PyTorch to JAX. Now, it’s running smoothly on TPUs! We still have challenges ahead, there is no good LoRA library in JAX, but this feels like the right path forward. Here's a demo (https://dub.sh/felafax-demo) of our managed solution. Would love your thoughts on our repo and vision as we keep chugging along!
Does the same job
all alternatives →
Predibase Reinforcement Fine-Tuning2025 · ▲175LLM reinforcement fine-tuning platform to improve LLM output
- LKLLMKube – Kubernetes for Local LLMs with GPU AccelerationNov 2025 · github.com · ▲5
Hi HN! I built LLMKube, a Kubernetes operator for deploying GPU-accelerated LLMs in production. One command gets you from zero to inference with full observability. Why this exists: Regulated industries (healthcare, defense, finance) need air-gapped LLM deployments, but existing tools are either single-node only (Ollama) or lack GPU optimization and SLO enforcement. LLMKube bridges the gap. What's working: - 17x speedup with NVIDIA GPUs (64 tok/s on Llama 3.2 3B vs 4.6 tok/s CPU) - One command: llmkube deploy llama-3b --gpu (auto CUDA setup, scheduling, layer offloading) -…
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, September 2024
the whole month →
BeforeSunset AI 2.0▲1,267Personalized AI daily planning that suits your life
AI · 2024 · beforesunset.ai
