Runpod Blog

While Oobabooga is a popular choice for text-based AI roleplay, KoboldAI offers a powerful alternative with smart context handling, more flexible editing, and better long-term memory retention. This guide compares the two frontends and walks through deploying KoboldAI on Runpod for writers and roleplayers looking for a deeper, more persistent AI interaction experience.

Oobabooga now supports up to 8192 tokens of context, up from the previous 2048-token limit. Learn how to upgrade your install, download compatible models, and optimize your setup to take full advantage of expanded memory capacity in longform text generation.

Runpod now offers access to NVIDIA’s powerful H100 GPUs, designed for generative AI workloads at scale. These next-gen GPUs deliver 7–12x performance gains over the A100, making them ideal for training massive models like GPT-4 or deploying demanding inference tasks.

Runpod's new Faster-Whisper endpoint delivers 2–4x faster transcription speeds than the original Whisper API—at a fraction of the cost. Perfect for podcasts, interviews, and multilingual speech recognition.

Want a custom spin on Stable Diffusion? This post shows you how to create and launch your own Vlad Diffusion template inside RunPod.

Oobabooga has a 2048-token context limit, but with the Long Term Memory extension, you can store and retrieve relevant memories across conversations. This guide shows how to install the plugin, use the Character panel for persistent memory, and work around current context limitations.

Learn how to install and use Bark AI on Runpod to generate realistic, expressive synthetic voices for narration, videos, or voiceover projects—no voice cloning required.