## Training with JAX

- [### Train Llama 3-8B using JAX, Ray, and GKE on Trillium
  Perform distributed training of the Llama 3-8B model on GKE using JAX, Ray Train, and TPU v6e (Trillium) with MaxText for optimized multi-host scaling.](https://cloud.google.com/kubernetes-engine/docs/tutorials/distributed-training-tpu)
- [### Pretrain Llama 3.1-70B using GKE clusters on Ironwood
  Train the Llama 3.1-70B model on TPU7x (Ironwood) using the MaxText framework.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/ironwood/llama3.1-70b)
- [### Pretrain DeepSeek 3-671B using GKE clusters on Ironwood
  Train the DeepSeek 3-671B model on TPU7x using optimized recipes for large-scale Mixture-of-Experts (MoE) architectures.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/ironwood/deepseek3-671b)
- [### Pretrain GPT OSS-120B using GKE clusters on Ironwood
  Train the GPT OSS-120B reasoning model on TPU7x using optimized recipes for large-scale distributed training.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/ironwood/gpt-oss-120b)
- [### Pretrain Qwen 3-235B using GKE clusters on Ironwood
  Train the Qwen 3-235B-A22B MoE model on TPU7x using optimized recipes for high-performance reasoning.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/ironwood/qwen3-235b-a22b)
- [### Pretrain Wan 2.1-14B using GKE clusters on Ironwood
  Train the Wan 2.1-14B video generation model on TPU7x using optimized recipes for high-performance video synthesis.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/ironwood/wan2.1-14b)
- [### Pretrain GPT3-175B using GKE clusters on Trillium
  Train the GPT3-175B model on TPU v6e using MaxText and optimized recipes for large-scale, cost-effective performance.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/trillium/GPT3-175B-MaxText)
- [### Pretrain Gemma3-12B using GKE clusters on Trillium
  Train the Gemma3-12B model on TPU v6e using MaxText and optimized recipes for high-performance open-model development.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/trillium/Gemma3-12B-MaxText)
- [### Pretrain Llama 3.1-70B using GKE clusters on Trillium
  Train Llama 3.1-70B on TPU v6e using MaxText and optimized recipes for high-throughput, large-scale model training.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/trillium/Llama3.1-70B-MaxText)
- [### Pretrain Llama 3.1-8B using GKE clusters on Trillium
  Train Llama 3.1-8B using MaxText on TPU v6e with this optimized recipe for scalable and high-performance pre-training.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/trillium/Llama3.1-8B-MaxText)
- [### Pretrain Mixtral-8x22B using GKE clusters on Trillium
  Train Mixtral-8x22B on TPU v6e using MaxText for optimized performance and efficiency.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/trillium/Mixtral-8x22B-MaxText)
- [### Pretrain Mixtral-8x7B using GKE clusters on Trillium
  Train Mixtral-8x7B using MaxText on TPU v6e with optimized configurations for high-throughput MoE performance on Google Cloud.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/trillium/Mixtral-8x7B-MaxText)
- [### Pretrain DeepSeek 3-671B using GKE clusters on v5p
  Train and deploy the DeepSeek 3-671B model on TPU v5p using MaxText for optimized large-scale performance.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/v5p/DeepSeek3-671B-MaxText)
- [### Pretrain GPT3-175B using GKE clusters on v5p
  Train the GPT3-175B model on TPU v5p using MaxText with optimized configurations for large-scale distributed training.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/v5p/GPT3-175B-MaxText)
- [### Pretrain Mixtral-8x7B using GKE clusters on v5p
  Train Mixtral-8x7B on TPU v5p using MaxText with optimized configurations for high-performance MoE workloads.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/v5p/Mixtral-8X7B-Maxtext)
- [### Pretrain SDXL using GKE clusters on v5p
  Train and scale Stable Diffusion XL (SDXL) on TPU v5p using MaxDiffusion for high-performance generative AI workloads.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/v5p/SDXL-MaxDiffusion)

## Training with PyTorch

- [### Pretrain Qwen 3-32B with TorchTitan using GKE clusters on Ironwood
  Train the Qwen 3-32B model on TPU7x (Ironwood) using PyTorch, TorchTPU, and the TorchTitan framework.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/ironwood/torchtitan/qwen3-32b/tpu7x-4x4x4/cluster_toolkit)
- [### Pretrain Qwen 3-30B-A3B with TorchTitan using GKE clusters on Ironwood
  Train the Qwen 3-30B-A3B Mixture-of-Experts (MoE) model on TPU7x using PyTorch, TorchTPU, and TorchTitan.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/ironwood/torchtitan/qwen3-30b-a3b/tpu7x-4x4x4/cluster_toolkit)
- [### Pretrain Qwen 3-8B with TorchTitan using GKE clusters on Ironwood
  Train the Qwen 3-8B model on TPU7x using PyTorch, TorchTPU, and TorchTitan for scalable distributed training.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/ironwood/torchtitan/qwen3-8b/tpu7x-2x2x2/cluster_toolkit)
- [### Pretrain Muse Glimmer 30B with TorchTitan using GKE clusters on Ironwood
  Train the Muse Glimmer 30B model on TPU7x using PyTorch, TorchTPU, and TorchTitan.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/training/ironwood/torchtitan/muse_glimmer_30b/tpu7x-4x4x4/cluster_toolkit)
- [### Pretrain Llama 3-70B with TorchTitan on Ironwood
  Train the Llama 3-70B model on TPU7x (Ironwood) using PyTorch, TorchTPU, and TorchTitan.](https://github.com/google-pytorch/torchtitan/tree/main/torchtitan/experiments/tpu/recipes/ironwood/llama3-70b/tpu7x-64)
- [### Pretrain Llama 3-8B with TorchTitan on Trillium
  Train the Llama 3-8B model on TPU v6e (Trillium) using PyTorch, TorchTPU, and TorchTitan.](https://github.com/google-pytorch/torchtitan/tree/main/torchtitan/experiments/tpu/recipes/trillium/llama3-8b/v6e-8)

## Inference with JAX

- [### Serve Llama 3.1-70B using GKE and vLLM on Trillium
  Serve LLMs on GKE using TPU v6e and vLLM, featuring optimized autoscaling and high-performance model serving on Google Cloud.](https://cloud.google.com/kubernetes-engine/docs/tutorials/serve-vllm-tpu)
- [### Serve LLMs using GKE with KubeRay
  Serve an LLM using TPUs on GKE with the Ray Operator add-on and the vLLM serving framework.](https://cloud.google.com/kubernetes-engine/docs/tutorials/serve-llm-tpu-ray)
- [### Serve open LLMs using GKE with Terraform
  Provision a GKE inference environment and deploy open LLMs using TPUs and a pre-configured Terraform architecture.](https://cloud.google.com/kubernetes-engine/docs/tutorials/serve-open-models-tpu-terraform)
- [### Serve Stable Diffusion XL (SDXL) using GKE
  Serve Stable Diffusion XL (SDXL) on GKE using Cloud TPUs and the MaxDiffusion framework for high-performance image generation.](https://cloud.google.com/kubernetes-engine/docs/tutorials/serve-sdxl-tpu)
- [### Serve GPT OSS-120B with vLLM using GKE clusters on Ironwood
  Run high-performance inference for GPT-OSS models on TPU7x using vLLM for optimized throughput and low-latency serving on Google Cloud.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/ironwood/vLLM/GPT-OSS)
- [### Serve Qwen3-Coder-480B with vLLM using GKE clusters on Ironwood
  Serve Qwen3-Coder-480B-A35B on TPU7x using vLLM for optimized, high-throughput code generation and inference.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/ironwood/vLLM/Qwen3-Coder-480B-A35B)
- [### Serve Llama 3.1-8B with v﻿LLM on Trillium
  Serve Llama 3.1-8B on TPU v6e using vLLM for optimized, low-latency inference and high-throughput serving.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/trillium/vLLM/Llama3.x)
- [### Serve Qwen 3 with vLLM on Trillium
  Serve Qwen 3 models on TPU v6e using vLLM for high-performance, scalable inference and optimized throughput.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/trillium/vLLM/Qwen3)
- [### Serve Qwen2.5-32B with vLLM on Trillium
  Serve the Qwen2.5-32B model on TPU v6e using vLLM for optimized, high-throughput inference.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/trillium/vLLM/Qwen2.5-32B)
- [### Serve Qwen2.5-VL with v﻿LLM on Trillium
  Serve Qwen2.5-VL vision-language models on TPU v6e using vLLM for optimized, high-performance multimodal inference.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/trillium/vLLM/Qwen2.5-VL)

## Inference with PyTorch

- [### Serve Qwen 3.5-397B with vLLM-TorchTPU using GKE clusters on Ironwood
  Serve the Qwen 3.5-397B model on TPU7x using PyTorch and vLLM-TorchTPU for multi-host distributed inference.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/ironwood/vLLM/vLLM-TorchTPU/Qwen3.5-397B/multi-host)
- [### Serve Gemma 4-26B with vLLM-TorchTPU on Ironwood
  Serve the Gemma 4-26B model on TPU7x using PyTorch and the vLLM-TorchTPU backend for high-throughput inference.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/ironwood/vLLM/vLLM-TorchTPU/Gemma4-26B)
- [### Serve GLM-5.2 with vLLM-TorchTPU on Ironwood
  Serve the GLM-5.2 model on TPU7x using PyTorch and the vLLM-TorchTPU backend.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/ironwood/vLLM/vLLM-TorchTPU/GLM-5.2)
- [### Serve Qwen 3-32B with vLLM-TorchTPU on Trillium
  Serve the Qwen 3-32B model on TPU v6e (Trillium) using PyTorch and the vLLM-TorchTPU backend.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/trillium/vLLM/vLLM-TorchTPU/Qwen3-32B/single-host)
- [### Serve Qwen 3.5-35B-A3B with vLLM-TorchTPU on Trillium
  Serve the Qwen 3.5-35B-A3B Mixture-of-Experts (MoE) model on TPU v6e using PyTorch and the vLLM-TorchTPU backend.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/trillium/vLLM/vLLM-TorchTPU/Qwen3.5-35B-A3B/single-host)
- [### Serve Gemma 4-26B with vLLM-TorchTPU on Trillium
  Serve the Gemma 4-26B model on TPU v6e using PyTorch and the vLLM-TorchTPU backend.](https://github.com/AI-Hypercomputer/tpu-recipes/tree/main/inference/trillium/vLLM/vLLM-TorchTPU/Gemma4-26B)