The compute platform for agents

Give your LLM access to GPUs, compute and sandboxes on the only platform designed for automating the full research loop.

Get early access

Divergent Labs gives the most ambitious teams the tools to build their own agentic machine learning lab. We give agents everything they need to research, build and test specialized models, from small classifiers to a fine-tuned large language model.

A powerful CLI

An interface for you and your agents: plain commands, plus skills any agent can pick up.

~/very-cool-classifier

# Run a training run in a sandboxed environment with a $5 budget
$ div run --gpu h100 train.py::run --max-spend 5.00

# Pull and push anything to super fast block storage
$ div storage get training-volume-1 log/training.jsonl

# Stream logs your agent can monitor and act upon in realtime
$ div logs training-run-4 -f

# Publish a model in a private repository, for your services to pull
$ div model publish very-cool-classifier/run-4

# Deploy inference endpoints in one go
$ div endpoint create very-cool-classifier/run-4 --gpu h100 --max-pool 4

Anyone can be a machine learning researcher

Our interfaces and skills are built for agents.

Agentbudget $2,000

Fine-tune a 70B open model on our 38k annotated contracts. It has to beat our frontier-model baseline on clause review. Budget is $2,000.

LoRA on 4 nodes of 8× H200, evaluating against your held-out set every 600 steps.

div storage putcontracts-v3 ./data✓ 38,412 docs

div run--gpu h200 --nodes 4 finetune.py::lora --max-spend 1500✓ $1,128

div evalrun-7 --suite clause-review✓ 87.2% vs 81.5%

Run 7 beats the baseline by 5.7 points. Eval loss bottomed out at step 5,400, so I kept that checkpoint.

div model publishcontracts-70b/run-7✓ private

Deploy it on 2× H200 behind an endpoint

contracts-70b · run-7done

traineval

$ div logs run-7 -f

step 4500 loss 0.522

step 4800 loss 0.559 eval 0.667

step 5100 loss 0.518

step 5400 loss 0.513 eval 0.657

step 5700 loss 0.512

step 6000 loss 0.545 eval 0.740

step6,000

best eval0.657

spend$1,128

Pricing

Pay for what your agents use, billed by the minute.

GPU

B300$8.45/h$0.1408/min
H200$5.45/h$0.0908/min
H100$4.45/h$0.0742/min

CPU

Per core$0.10/core/h$0.0017/core/min

Other

Logs$0.30/GB ingested
Block storage$0.10/GB/month

Custom

Design a custom package – available for businesses with ambitious projects, startups or large volume.

Talk to us

10% off CPUs and GPUs on weekends1.10× surcharge in the EEA

Give your agents a lab of their own.

Get early access