SaladCloud Blog
All Tags
Earning Rate Changes: September 12
Salad earning rates change September 12, 2026: RTX 5090 and 4090 rates go up, 15 GPU classes come down at higher priority tiers, CPU payouts rise.
Upcoming SaladCloud Price Changes: September 12, 2026
SaladCloud GPU and vCPU pricing is changing on September 12, 2026: 15 GPU classes drop 25% on average, while RTX 5090 and 4090 prices increase.
How SaladCloud Built a Global Distributed Compute Network
See how SaladCloud built a global distributed compute network across 191 countries by turning idle consumer GPUs into cloud infrastructure.
Can Idle GPUs Help Reduce AI's Carbon Footprint?
Learn how SaladCloud reuses idle GPUs to help reduce AI's carbon footprint by getting more compute from hardware that already exists.
SaladCloud Recipes: The Fastest Way to Deploy on Distributed GPUs
Learn how SaladCloud recipes simplify distributed GPU deployment with tested defaults, reusable templates, and ready-to-run configurations.
Bringing Salad's Payment Rails On-Chain: Our Milestone 3 Architecture
Salad's implementation design for RNP-023 Milestone 3: x402 batch settlement, daily on-chain burns, and RENDER Chef rewards.
Salad x Render Network: Milestones 1 and 2 Are Live
RNP-023 Approved: Salad Is Joining the Render Network
Salad is joining the Render Network after RNP-023 approval. Learn what this means for chefs, customers, and the future of decentralized GPU compute.
Use Cline with SaladCloud: Building Real Apps for Under $0.01
See how to use Cline with a SaladCloud-hosted Qwen 3.5 model to build real apps, including web pages, browser games, and streamlit dashboards starting under $0.01 total.
Salad Proposes Integration with the Render Network
Salad has submitted a formal proposal to become a subnet on the Render Network, transitioning payments and rewards to on-chain RENDER transactions. Learn what this means for customers and Chefs.
Qwen 3.5 Small Models on SaladCloud: Benchmarks, Cost, and Why You Don't Need a Mac Mini
Benchmarking Qwen 3.5 small models on SaladCloud: performance, cost, throughput, and why cloud inference can beat buying a Mac Mini.
Building an Automated News Bot with OpenClaw and SaladCloud: A Real-World Cost Breakdown
Learn how to build an automated news bot with OpenClaw and SaladCloud, reduce LLM costs by 86% with $0.09/hour GPU.
Reduce Your OpenClaw LLM Costs: SaladCloud Guide
Running OpenClaw can get expensive fast. Learn how to cut LLM costs by self-hosting models on SaladCloud with flat hourly GPU pricing.
The Future of Scientific Compute is Decentralized
Scientific computing is shifting from supercomputers to decentralized GPU networks cutting costs by over 85% and unlocking planetary-scale research.
Giving AI Agents Browser Access at the Edge
MCProxy gives AI agents browser access from residential IPs worldwide—bypassing bot detection for less than $0.01/hr per location.
Zero-Knowledge Proof Mining at Scale with SWPS & SaladCloud
How SWPS and SaladCloud power zero-knowledge proof mining at scale using distributed GPU compute, enabling affordable and decentralized ZKPoW infrastructure.
Nockchain ZK Proving on SaladCloud
Learn how Nockchain's zero-knowledge proof mining works on SaladCloud with GPU compute infrastructure optimized for web3 workloads. Discover setup insights, performance benefits, and how to deploy ZK proof miners at scale.
Save up to 77% on Zero-Knowledge Proofs with SaladCloud
Prove more, spend less: decentralized GPUs on SaladCloud for ZKPs. Fast runs, predictable costs, and simple scaling for modern zk stacks.
Hunyuan3D 2.1 Image-to-3D: Achieve $0.009/generation on SaladCloud and save 94% over FAL
Image-to-3D up to 94% cheaper than FAL with SaladCloud RTX 4090s.
Benchmarking SaladCloud: GPU Performance and Reliability in Action
SaladCloud deployed a workload with 100 replicas over seven days to measure key metrics related to performance and reliability.
Salad Slashes GPU Pricing Again – Now the Serving the Lowest Cost GPUs on the Market
SaladCloud announces the most affordable cloud computing platform with the lowest cost GPUs for H100s, A100s, RTX 4090s, and more.
SaladCloud Logging Updates - Better Reliability, Structured Logs, and more
The SaladCloud logging includes new features for a log query API and rock-solid reliability. Learn more more from our engineering team.
FABLE: Fiction Adapted BERT for Literary Entities
Salad releases our first open model, FABLE, a named-entity recognition model based on DeBERTa v3 specializing in narrative fiction.
Generate a Minute of High-Quality AI Video with WAN2.2 on SaladCloud for as Little as $0.87
WAN2.2 is the best open source video model yet. Learn how to run this on SaladCloud GPUs for the best cost per minute of finished video.
Blender Rendering Benchmark across 27 Consumer and Data Center GPUs on SaladCloud, with 90%+ Cost Savings
Blender is equipped with industry-grade tools for modeling, animation, and rendering. Learn more how to run this workload on SaladCloud.
Benchmarking WAN2.1: Open-Source Text-to-Video Generation on SaladCloud
Read more about the WAN2.1 benchmark running on Salad consumer vs. enterprise GPU cards, the most cost effective solution on the market.
Building High-Availability Services on Salad
Building reliable applications on distributed, interruptible infrastructure is a solved engineering problem. Achieve 50-85% cost savings now.
New SaladCloud Recipes: Ready-to-Deploy Solutions for Your Distributed Computing Needs
Looking for ready-to-deploy application solutions? Find new SaladCloud recipes for molecular dynamic simulations, model training, and more.
Redefining AI Infrastructure: How TheStage AI and Salad Achieved 10K+ Images Per Dollar Through Distributed Optimization
Learn more how TheStage AI and Salad achieved 10k+ images per dollar through distributed optimization and Flux.1-Schnell.
Welcome to the Salad Kitchen!
Learning how to deploy on Salad? Visit the Salad Developer Hub to review tutorials, how-to-guides, model recipes, and more.
How Sogni Leveraged Salad's Distributed GPU Network to Scale from 3.5K to 340K+ Users in 90 Days
Salad's Distributed GPU Network helped Sogni scale from 3.5k to over 340k users in 90 days. Leading to cost savings and affordable access.
Introducing Salad's Transcription API: superior accuracy now available at 40% of the cost of other providers
Salad Transcription API is the most accurate AI batch transcription API with 95.1% accuracy for English, and at 40-80% less cost than others.
Salad Transcription API Accuracy Benchmark: How it outperforms Deepgram, Assembly AI & AWS
In this blog, we share results from an accuracy benchmark of Salad Transcription API, scoring 95.1% accuracy rate - No.1 in the industry.
Project Odyssey: Generative AI Empowers Independent Creators
Project Odyssey shows off the best generative AI has to offer - closed and open source.
Dirt Cheap Image Captioning With Qwen Vision-Language Models: Up to 98.4% Cheaper than OpenAI
In this benchmark, we compare Qwen image captioning against OpenAI, showing a 98.4% cost reduction with Qwen open vision language model.
There's An AI For That scales user-generated AI tools with SaladCloud at 50% less cost
There's An AI For That (TAAFT), an AI tools platform, scales 100s of user-generated AI tools on SaladCloud at 50% less cost.
SaladCloud becomes the first to offer NVIDIA RTX 5090/5080s, thanks to the power of community
SaladCloud becomes the first provider to offer the RTX 5090/5080 GPUs for rent, starting from just $0.195 per hour.
DeepSeek TGI Benchmark: Build highly scalable and cost-effective LLM applications on SaladCloud, at 70% less cost than hyperscalers
In this article, we share a DeepSeek TGI benchmark on SaladCloud, showing how to save up to 70% on compute cost compared to hyperscalers.
1000s of GPUs, over 50% savings: Klyne accelerates AI drug discovery on SaladCloud
"With SaladCloud, we can scale up to 1000s of GPUs quickly at the lowest cost" - Klyne accelerates AI drug discovery with SaladCloud.
How Undetectable.ai saves over $50,000 a month with SaladCloud's consumer GPUs
Undetectable AI chose SaladCloud's consumer GPUs over A100s on a hyperscaler, saving $50k a month on LLM inference GPUs.
Stable Diffusion XL (SDXL) benchmark: 3405 images per dollar on SaladCloud
This SDXL benchmark on SaladCloud shows cost-effective and scalable AI image generation on SaladCloud, delivering 3405 images per dollar.
Stable diffusion 1.5 benchmark: 14,000+ images per dollar on SaladCloud
In this benchmark, we evaluate stable diffusion 1.5 on 3 consumer GPUs, with the RTX 3060 Ti offering the best price performance.
Molecular Simulation: GROMACS Benchmark on 30 GPUs on SaladCloud, 90+% Cost Savings
We benchmark GROMACS on 30 GPUs on SaladCloud, achieving a 90% cost savings by using consumer GPUs for molecular simulations.
Salad drops GPU pricing to become the lowest priced cloud (again)
Salad Technologies reduces price on all GPUs, offering users the lowest Cloud GPU pricing in the market - with 450,000+ compute providers.
Molecular Simulation: OpenMM Benchmark on 25 Consumer GPUs, 95% Less Cost
How to Run Cog Applications on SaladCloud
Overpaying for GPUs? In this blog, we show how to run cog applications on Salad to switch your AI workloads to SaladCloud.
Blend cuts AI inference cost by 85% on SaladCloud while running 3X more scale
Civitai powers 10 Million AI images per day with Salad's distributed cloud
By switching to SaladCloud, Civitai serves inference on 600+ GPUs, delivering 10 Million images per day & training 15,000+ LoRAs per month.
AI Batch Transcription Benchmark: Transcribing 1 Million+ Hours of Videos in just 7 days for $1800
In this AI batch transcription benchmark, we transcribed 1 Million hrs of YouTube videos for $1800 on SaladCloud (with Parakeet TDT 1.1B).
Introducing SSAP: Migrate to SaladCloud easily & save up to 80%
The Salad Solutions Architect Program (SSAP) helps AI companies struggling with high cloud/inference cost to switch to Salad GPU Cloud easily.
AI Transcription Benchmark: 1 Million Hours of Youtube Videos with Parakeet TDT 1.1B for Just $1260, a 1000-fold cost reduction
In this benchmark, we transcribe 1 Million hours of YouTube videos in a batch processing mode on SaladCloud for just $1260, a 1000-fold cost reduction.
Text-to-Speech (TTS) API Alternative: Self-Managed OpenVoice vs MetaVoice Comparison
We compare OpenVoice and MetaVoice, showing how self-managed open-source models can deliver ~90% cost savings over text to speech API.
Fine-tuning Stable Diffusion with Dreambooth on SaladCloud (from just $0.1426 per training job)
In this tutorial, we share how to perform Stable Diffusion fine tuning with Dreambooth on SaladCloud from just $0.1426 per training job.
MetaVoice AI Text-to-Speech (TTS) Benchmark: Narrate 100,000 words for only $4.29 on SaladCloud
In this MetaVoice-1B Text-to-Speech benchmark, we compare cost and speed for converting 100,000 words to speech for different GPUs on Salad.
Fine-Tuning Stable Diffusion XL (SDXL) with interruptible GPUs and LoRA for low cost
In this tutorial, we share how to fine tune Stable Diffusion XL (SDXL) with interruptible GPUs and LORA for a low cost using consumer GPUs.
Parakeet TDT 1.1B Inference Benchmark on SaladCloud: 1,000,000 hours of transcription for Just $1260
Inference Benchmark on SaladCloud: Distil-Whisper Large V2 vs. Whisper Large V3 for Speech-to-text
In this speech-to-text inference benchmark, we compare Distil-Whisper Large V2 and Whisper-Large V3 across different clouds and managed APIs.
OpenVoice Text-to-Speech (TTS) Benchmark: 6 Million+ Words/$ Using SaladCloud
In this GPU benchmark for OpenVoice on SaladCloud, consumer GPUs deliver 6 Million+ words per dollar for Text-to-Speech inference.
Whisper Large V3 Speech Recognition Benchmark: 1 Million hours of audio transcription for just $5110
In this Whisper Large v3 benchmark, we show how AI companies can deliver transcription at a 99% lower cost compared to managed services.
Tag 309K Images/$ with Recognize Anything Model++ (RAM++) On Consumer GPUs
In this benchmark of the Recognize Anything Model++ (RAM++), the RTX 2080 tagged 309,000 images per dollar on SaladCloud's consumer GPUs.
Tutorial: How to run your own GPU-accelerated JupyterLab on SaladCloud
In this tutorial, you will learn how to deploy and run your own GPU-accelerated JupyterLab on Salad's network of 10K+ affordable GPUs.
Segment Anything Model (SAM) Benchmark: 50K Images/$ on Consumer GPUs
In this benchmark of Segment Anything Model (SAM), the RTX 3060 Ti and RTX 3070 Ti offer 50X the cost efficiency of other managed services.
Your own ChatGPT for just $0.04/hr - with Ollama, ChatUI and SaladCloud
We explore how to build your own ChatGPT with Ollama, Huggingface Chat UI and SaladCloud for just $0.04 per hour.
LLM Comparison Through TGI Benchmark Using SaladCloud
In this LLM comparison, we compare 4 popular large language models (Mistral-7b, CodeLlama, Falcon-7b, santacoder) through TGI on SaladCloud.
Data Pipeline Processing with GPUs: Why, How, and Where
Data pipeline processing is a bottleneck for today's AI industry due to high GPU prices and shortages. Distributed clouds are the solution.
Analyzing the Stunning Realism of GTA6 with YOLOv8 and SaladCloud
We use the YOLOv8 object detection model on SaladCloud to analyze the popular GTA6 trailer and count the objects in the trailer.
Training a custom YOLOv8 model on SaladCloud for just $0.25
In this blog, we showcase training three distinct custom YOLOv8 models on SaladCloud within an hour for just $1.
Stable Diffusion v1.5 Benchmark On Consumer GPUs
In this Stable Diffusion Benchmark, we compare SD v1.5 across 23 consumer GPUs generating more than 460,000 QR codes on SaladCloud.
YOLOv8 Benchmark: Object Detection on SaladCloud's GPUs (73% Cheaper Than Azure)
In this benchmark, we deploy YOLOv8 on GPUs across Azure & SaladCloud. Salad is 73% cheaper for object detection using YOLOv8.
Announcing Salad's SOC 2 Type 1 Certification
SaladCloud Launches Kubernetes Support at KubeCon 2023
Salad launches Kubernetes support for its GPU cloud at KubeCon 2023 with the SaladCloud Virtual Kubelet Provider.
Comparing Price-Performance of 22 GPUs for AI Image Tagging (GTX vs RTX)
In this AI image tagging benchmark, we compare 22 GPUs. Find out which GTX-series GPU gets 3X more images/$ than a popular RTX-series GPU.
Bark Benchmark: Reading 144K Recipes with Text-to-Speech on SaladCloud
In this benchmark of suno-ai/bark, we read 144K recipes with bark's text-to-speech model on Salad's consumer GPUs.
The AI GPU Shortage: How Gaming PCs Offer a Solution and a Challenge
Distributed clouds utilizing latent gaming PCs offer cost-efficiency with reliability in a transformative way to tackle the AI GPU shortage.
Stable Diffusion XL (SDXL) Benchmark - 769 Images Per Dollar on SaladCloud
In this Stable Diffusion XL (SDXL) benchmark, SaladCloud delivered 769 images per dollar, running on RTX 3090 and 4090 GPUs.
Whisper Large Inference Benchmark: 137 Days of Audio Transcribed in 15 Hours for Just $117
In this inference benchmark for whisper-large-v2 model, we transcribed 137 days of audio in 15 hours for just $117 on SaladCloud.
Exploring AI Bias by Turning Faces into Salads - An Experiment
We explore AI bias in image generation by building our own portrait generator in the style of famous Italian painter - Giuseppe Arcimboldo.
A New Price-Performance Standard for BERT Transformers.
Engineers from Numenta used SaladCloud to get 10x more inferences per dollar for BERT transformer workloads.
How Dreamup reduces AI image generation cloud cost with SaladCloud
"Salad's GPU pricing is much better than everywhere else" - DreamUp, a Stable Diffusion based image generator switched to Salad to reduce cloud costs.
Stable Diffusion Inference Benchmark
In this Stable Diffusion GPU benchmark for inference, we generated 9.2 Million images in 24 hours on 750 consumer-grade GPUs- all for $1872.
2022 In Review
2022 was a phenomenal year for Salad chefs, customers and partners. Over 494,000 chefs joined Salad's mission to democratize the cloud.
Centralization Is a Systems Design Problem
In this blog, we discuss how today's centralized cloud is really a monopoly and the role of distributed infrastructure in breaking that.
Decentralize, or Be Destroyed!
We can decrease the reliance on centralized solutions by supporting alternative compute applications like decentralized/distributed clouds.
Perfect Device Fingerprinting at Scale
How we achieved a 100% web scraping success rate on SaladCloud - without any attempt to spoof device fingerprinting.
Exploring Distributed Data Parallel (DDP) Training on Consumer GPUs
Salad's distributed cloud has 10k+ active nodes at any time, enabling applications like distributed data parallelism (DDP).
Government Science on Citizen Machines - Why Distributed Cloud is the Way
Cloud computing is a monopoly today. Distributed cloud computing is our best way to fight back against soaring prices & long wait-times.
How the Kitchen Conquered 2021
A review of 2021 for Salad - 60,000 charitable donations, giveaways and 950% more people joining Salad's mission to democratize the cloud.
How to Refactor Public Cloud to Reduce Cloud Cost
$397.5B in end-user spending. $206B in profits for cloud providers - a cozy 51% margin. It's time for a leaner cloud and lower cloud cost.
SaladCast Episode 10: The Salad Business Model With CEO Bob Miles
In this episode, we chat with Bob Miles, CEO of Salad, on our business model and how we connect unused GPUs to innovative companies.
Salad Raises $17M to Decentralize the Cloud
Thousands of PC gamers and an incredible lineup of investors are joining us in our mission to democratize the cloud.
SaladCast Episode 12: Our Product Roadmap With Daniel Sarfati
In this episode, we discuss Salad's roadmap, building a distributed cloud and diversified workloads with our VP of Product.
SaladCast Episode 13: Kyle Dodson on Containerized Workloads
Engineering mastermind Kyle Dodson fields some heady questions about the history of containerization technology.
SaladCast Episode 11: Jared Carpenter on Salad's Go-to-Market Strategy
In this episode, we discuss Salad's acquisition strategy to introduce gamers into our ecosystem and help them earn rewards with their PC.
SCE Beta Registration Now Open
Salad Container Engine beta registration is now open. Switch to SaladCloud and save up to 90% on your cloud cost.
Solving for Cost: Why Innovators Need a Decentralized Cloud
In today's age of GPU shortages, long wait times and high prices, a decentralized cloud is the answer to AI innovators' appetite for compute.
Stable Diffusion v1.4 Inference Benchmark - GPUs & Clouds Compared
In this gpu benchmark, we evaluate the inference performance of Stable Diffusion 1.4 on different compute clouds and GPUs.
GPU shortage isn't the problem for Generative AI. GPU selection is.
Waiting for expensive A100s & H100s for serving AI inference at scale? AI GPU shortage isn't the problem. GPU selection is.
Optimizing AI Image Generation: Streamlining Stable Diffusion with ControlNet in Containerized Environments
In this blog, we show how to optimize controlnet implementation for stable diffusion in a containerized environment on SaladCloud.