Pricing

Documentation

Blog

Case Studies

Get Access

Batch LLM Inference is better with Sutro

Run LLM Batch Jobs in Hours, Not Days, at a Fraction of the Cost.

Get Started

Generate a question/answer pair for the following chunk of vLLM documentation

Inputs

Outputs

Intro to vLLM

vLLM is a fast and easy-to-use library for LLM inference and serving. Originally developed in the Sky Computing Lab at UC Berkeley, vLLM has evolved into a community-driven project with contributions from both academia and industry.

Loading Models

vLM models can be loaded in two different ways. To pass a loaded model into the vLLM framework for further processing and inference without reloading it from disk or a model hub, first start by generating


Using the Open AI Server

Run:ai Model Streamer is a library to read tensors in concurrency, while streaming it to GPU memory. Further reading can be found in Run:ai Model Streamer Documentation.

vLLM supports loading weights in Safetensors format using the Run:ai Model Streamer. You first need to install vLLM RunAI optional dependency:

Question: Is vLLM compatible with all open-source models? ...

Question: How do I load a custom model from HuggingFace? ...

Question: Can I use the OpenAI compatible server to replace calls...

+128 more…

Batch LLM Inference is better with Sutro

Run LLM Batch Jobs in Hours, Not Days, at a Fraction of the Cost.

Get Started

Generate a question/answer pair for the following chunk of vLLM documentation

Inputs

Outputs

Intro to vLLM

vLLM is a fast and easy-to-use library for LLM inference and serving. Originally developed in the Sky Computing Lab at UC Berkeley, vLLM has evolved into a community-driven project with contributions from both academia and industry.

Loading Models

vLM models can be loaded in two different ways. To pass a loaded model into the vLLM framework for further processing and inference without reloading it from disk or a model hub, first start by generating


Using the Open AI Server

Run:ai Model Streamer is a library to read tensors in concurrency, while streaming it to GPU memory. Further reading can be found in Run:ai Model Streamer Documentation.

vLLM supports loading weights in Safetensors format using the Run:ai Model Streamer. You first need to install vLLM RunAI optional dependency:

Question: Is vLLM compatible with all open-source models? ...

Question: How do I load a custom model from HuggingFace? ...

Question: Can I use the OpenAI compatible server to replace calls...

+128 more…

LLM performance evaluation

Benchmark models in hours, not days

Benchmark your LLM outputs to continuously improve workflows, agents, and assistants. Easily evaluate custom models against any use case at a fraction of the cost.

Get Access

Generate a question/answer pair for the following chunk of vLLM documentation

Inputs

Outputs

Intro to vLLM

vLLM is a fast and easy-to-use library for LLM inference and serving. Originally developed in the Sky Computing Lab at UC Berkeley, vLLM has evolved into a community-driven project with contributions from both academia and industry.

Loading Models

vLM models can be loaded in two different ways. To pass a loaded model into the vLLM framework for further processing and inference without reloading it from disk or a model hub, first start by generating


Using the Open AI Server

Run:ai Model Streamer is a library to read tensors in concurrency, while streaming it to GPU memory. Further reading can be found in Run:ai Model Streamer Documentation.

vLLM supports loading weights in Safetensors format using the Run:ai Model Streamer. You first need to install vLLM RunAI optional dependency:

Question: Is vLLM compatible with all open-source models? ...

Question: How do I load a custom model from HuggingFace? ...

Question: Can I use the OpenAI compatible server to replace calls...

+128 more…

Batch LLM Inference is better with Sutro

Run LLM Batch Jobs in Hours, Not Days, at a Fraction of the Cost.

Get Started

Generate a question/answer pair for the following chunk of vLLM documentation

Inputs

Outputs

Intro to vLLM

vLLM is a fast and easy-to-use library for LLM inference and serving. Originally developed in the Sky Computing Lab at UC Berkeley, vLLM has evolved into a community-driven project with contributions from both academia and industry.

Loading Models

vLM models can be loaded in two different ways. To pass a loaded model into the vLLM framework for further processing and inference without reloading it from disk or a model hub, first start by generating


Using the Open AI Server

Run:ai Model Streamer is a library to read tensors in concurrency, while streaming it to GPU memory. Further reading can be found in Run:ai Model Streamer Documentation.

vLLM supports loading weights in Safetensors format using the Run:ai Model Streamer. You first need to install vLLM RunAI optional dependency:

Question: Is vLLM compatible with all open-source models? ...

Question: How do I load a custom model from HuggingFace? ...

Question: Can I use the OpenAI compatible server to replace calls...

+128 more…

Batch LLM Inference is better with Sutro

Run LLM Batch Jobs in Hours, Not Days, at a Fraction of the Cost.

Get Started

Generate a question/answer pair for the following chunk of vLLM documentation

Inputs

Outputs

Intro to vLLM

vLLM is a fast and easy-to-use library for LLM inference and serving. Originally developed in the Sky Computing Lab at UC Berkeley, vLLM has evolved into a community-driven project with contributions from both academia and industry.

Loading Models

vLM models can be loaded in two different ways. To pass a loaded model into the vLLM framework for further processing and inference without reloading it from disk or a model hub, first start by generating


Using the Open AI Server

Run:ai Model Streamer is a library to read tensors in concurrency, while streaming it to GPU memory. Further reading can be found in Run:ai Model Streamer Documentation.

vLLM supports loading weights in Safetensors format using the Run:ai Model Streamer. You first need to install vLLM RunAI optional dependency:

Question: Is vLLM compatible with all open-source models? ...

Question: How do I load a custom model from HuggingFace? ...

Question: Can I use the OpenAI compatible server to replace calls...

+128 more…

From Prototype to Production, Simplified

Sutro takes the pain away from testing and scaling LLM batch jobs to unblock your most ambitious AI projects.

import sutro as so

from pydantic import BaseModel

class ReviewClassifier(BaseModel):

sentiment: str

user_reviews = '.

User_reviews.csv

User_reviews-1.csv

User_reviews-2.csv

User_reviews-3.csv

system_prompt = 'Classify the review as positive, neutral, or negative.'

results = so.infer(user_reviews, system_prompt, output_schema=ReviewClassifier)

Progress: 1% | 1/514,879 | Input tokens processed: 0.41m, Tokens generated: 591k

█░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░

Prototype

Start small and iterate fast on your evaluation workflows. Accelerate experiments by testing on Sutro before committing to large jobs.

Scale

Scale your LLM evaluation workflows so your team can do more in less time. Process billions of tokens in hours, not days, with no infrastructure headaches or exploding costs.

Integrate

Seamlessly connect Sutro to your existing LLM workflows. Sutro's Python SDK is compatible with popular data orchestration tools, like Airflow and Dagster.

Benchmark at scale

Confidently evaluate models against millions of requests and billions of tokens at a time without the pain of managing infrastructure.

Reduce evaluation costs

Reduce evaluation costs

Reduce evaluation costs

Get benchmark results faster and reduce costs by 10x or more by parallelizing your LLM calls through Sutro.

Get feedback faster

Shorten development cycles by getting feedback from large evaluation jobs in as little as minutes before scaling up.

Synthetic data generation

Longer description goes here, should span multiple lines.

RAG data preparation

Easily convert large corpuses of free-form text into vector representations for semantic search.

Sentiment analysis

Easily sift through thousands of product reviews and unlock valuable product insights.

Structured Extraction

Transform unstructured data into structured insights that drive business decisions.

Embedding Generation

Easily convert large corpuses of free-form text into vector representations for semantic search and recommendations.

Document summarization

Convert massive amounts of free-form text into analytics-ready datasets without the pains of managing your own infrastructure.

Synthetic data generation

Longer description goes here, should span multiple lines.

RAG data preparation

Easily convert large corpuses of free-form text into vector representations for semantic search.

Sentiment analysis

Easily sift through thousands of product reviews and unlock valuable product insights.

Structured Extraction

Transform unstructured data into structured insights that drive business decisions.

Embedding Generation

Easily convert large corpuses of free-form text into vector representations for semantic search and recommendations.

Document summarization

Convert massive amounts of free-form text into analytics-ready datasets without the pains of managing your own infrastructure.

Synthetic data generation

Longer description goes here, should span multiple lines.

RAG data preparation

Easily convert large corpuses of free-form text into vector representations for semantic search.

Sentiment analysis

Easily sift through thousands of product reviews and unlock valuable product insights.

Structured Extraction

Transform unstructured data into structured insights that drive business decisions.

Embedding Generation

Easily convert large corpuses of free-form text into vector representations for semantic search and recommendations.

Document summarization

Convert massive amounts of free-form text into analytics-ready datasets without the pains of managing your own infrastructure.

FAQ

What are Sutro's core capabilities?

How does Sutro reduce costs?

How does Sutro handle large-scale jobs?

Does Sutro integrate with other tools?

How difficult is it to use Sutro?

What Will You Scale with Sutro?

Get Access

Blog

Privacy Policy

Terms of Service

Documentation

Docs

team@sutro.sh