Featherless Blog

Fine-Tuned Conversations II - Mastering the Art of LLM Customization with Auri

For our second interview in the series Fine-Tuned Conversations we speak with Auri, model creator of Celeste and the EVA series

Experimental support for Kimi-K2 by Moonshot AI now available for premium users!

Kimi-K2 is a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters.

Unleash 10,000+ Models in Dify: Introducing the Featherless.ai plugin

Build powerful AI applications with serverless inference and low-code orchestration

Prompt engineering and Context Engineering: The Complete Developer’s guide to Modern AI System Design

From crafting prompts to architecting intelligent information ecosystems

Building Resilient AI Applications with Featherless and LiteLLM: Fallbacks and Load Balancing

Showcasing the new Featherless integration with LiteLLM to build more resilient systems through intelligent fallbacks and efficient load balancing.

Fine-Tuned Conversations - Building Sawalni: How Omar is Creating AI for Moroccan Darija

In the first part of our series Fine-Tuned Conversations we have an interview with Omar, creator of Sawalni, sharing his journey from childhood coder to AI innovator

LLM Observability: How Featherless + OpenLIT Simplifies AI Monitoring.

Log chats, prompts and completions using OpenLIT

Featherless Becomes Hugging Face’s Largest LLM Inference Provider with 6,700+ Models

Unlimited token pricing and instant access now power the future of open-source AI deployment.

Context Isn’t Everything: Build Efficient LLM Apps with LlamaIndex + Featherless

Build Production RAG Applications with 4,000+ Open Source Models and Zero Infrastructure Overhead

Building Production-Ready LLM Apps with LangChain & Featherless Serverless Inference

LangChain Integration for Serverless LLM Deployment with 4,300+ Open Source Models

Qwen 3 & GLM 4: Next-Generation Thinking Models Now on Featherless.ai

Announcing Qwen 3 and GLM, two powerful families of advanced large language models, are now available through Featherless.ai's serverless inference platform.

Choose Your AI Muse: Connect Mikupad & Novelcrafter to Any Model via Featherless.ai

A practical guide to integrating Featherless.ai's powerful, flexible model inference with your Mikupad and Novelcrafter workflows.

Supercharging Your Development Workflow part 2: Integrating Featherless.ai with Cline and Roo Code

Integrating Open-Source VS Code Extensions Cline & Roo Code with featherless.ai

Running OpenHands LM 32B with Featherless.ai: A Practical Guide

Getting the best out of OpenHands with Featherless.ai

Initial Support for Google's Gemma 3 27B Models Now Live on Featherless.ai!

Gemma 3 27B model support and updates

Supercharging Your Development Workflow: Integrating Featherless.ai with Aider and Cursor

Connect DeepSeek V3, Qwen2.5-Coder , Openhands LM and 4000+ Open Models to Aider and Cursor Without Infrastructure Headaches

Featherless.ai Raises $5M in Seed Funding to Democratize Access to Open Source AI Models

Democratizing AI with instant access to 4,000+ open-source models through our breakthrough serverless platform

Building a PDF-to-Podcast Pipeline with Open-Source AI: From Text Extraction to Voice Synthesis

QwQ-32B Now Available on Featherless.ai

QwQ-32B: A Powerful Lightweight in the Age of Reasoning Models

DeepSeek-R1 available for premium users!

DeepSeek-R1 achieves performance comparable to OpenAI o1 across math, code and reasoning tasks. A SOTA open-source model with open reasoning tokens.

Zero to AI: Deploying Language Models Without the Infrastructure Headache

Where open-source models take flight: Your launchpad to production-ready AI

Running Open Source LLMs in Popular AI Clients with Featherless: A Complete Guide

A step-by-step walkthrough for developers and AI enthusiasts to integrate any Hugging Face model into their favorite chat client.

The hidden costs of local LLM inference

Drop the "thousand"; just "ten dollars" - it's cleaner.

I compared QwQ-32B-Preview to Marco-01

And here are the results

Featherless Feud: a dip into LLM-powered game development

rebuilding a TV classic

Model Support Summary + new this week

Nemo 12B and Qwen2 32B latest additions for a total of 12 model families

Featherless: an introduction

making every hugging face model available for inference and why it matters