LAUNCH OFFER — $5 free credits for new accounts · no card required
EX5.ai
PART OF THE EX5 ECOSYSTEM

Rent GPUs.
Run any model.
Ship AI.

ex5.ai is a GPU marketplace, a unified model API, a fine-tuning pipeline, a chat workspace and a coding CLI — one credit balance powers all of it. Top up with card.

Open console →
$npm i -g @ex5/cli
$curl api.ex5.ai/v1/chat/completions -d '{"model": "ex1-72b", ...}'
GPUs online now
starting price
models on the API
API uptime, 90 days
01 / COMPUTE
GPU Cloud

Bare-metal GPUs from verified hosts, billed per second — H200 down to RTX 4090.

02 / INFERENCE
Model API

One OpenAI-compatible endpoint for EX-1 and 50+ open models. Pay per token.

03 / TRAINING
Fine-tune & Deploy

Upload a dataset, tune EX-1 or any open model, get a private endpoint — one command.

04 / WORKSPACE
chat.ex5.ai + CLI

A fast web workspace for every model, and an agentic CLI for your terminal.

// GPU MARKET

Live GPU market

Real-time on-demand and interruptible pricing across regions. Rent in seconds, pay per second, cancel anytime.

Browse all offers →
GPU
VRAM
TFLOPS FP16
REGION
AVAIL
INTERRUPTIBLE
ON-DEMAND

prices refresh every 60s · all hosts verified & benchmarked · datacenter tier only

// MODEL API

Every model, one API

Swap the base URL and keep your code. Our flagship EX-1 plus the best open models, priced per million tokens and billed from the same balance as your GPUs.

curl
curl https://api.ex5.ai/v1/chat/completions \
  -H "Authorization: Bearer $EX5_KEY" \
  -d '{ "model": "ex1-72b", "messages": [...] }'
OpenAI-compatible streaming tool use
// SPECIALIZED MODELS

Specialized, not general-purpose

Task-tuned models we train in-house — smaller, faster and cheaper than a frontier model on their own turf. Every one has a playground in chat.ex5.ai and a stable API id.

// FINE-TUNE & DEPLOY

Your data in.
Your endpoint out.

No cluster setup, no training scripts. Fine-tune EX-1 or any open model on marketplace GPUs at spot prices, and we deploy the result as a private OpenAI-compatible endpoint the moment training finishes.

  1. 01
    Upload a JSONL
    Drag a dataset into the console or push it from the CLI — we validate, split and tokenize it.
  2. 02
    One command to train
    LoRA or full fine-tune with live loss curves. Cost is estimated up front and billed per second.
  3. 03
    Auto-deploy on finish
    A private, autoscaling endpoint on api.ex5.ai — or download the weights and take them anywhere.
Start a fine-tune →$ ex5 tune start --base ex1-72b --data ./support.jsonl
ft-2841RUNNING
2× H100 SXM · spot
BASE
ex1-72b
DATASET
support.jsonl
METHOD
LoRA r=32
epoch 2/3 · step 1,840/2,88064%
TRAIN LOSS
loss 0.412 ↓est. total $21.40 · elapsed 1h 12m
✓ deployed → https://api.ex5.ai/v1
"model": "acme-support-ex1" · autoscale 0→8 · weights downloadable
// CHAT.EX5.AI

A workspace for every model

Talk to EX-1, your fine-tunes or any catalog model — files, code and web search included. Metered per token, billed from the same balance.

Open the live demo →
chat.ex5.ai
ex1-72b ▾128K ctx · $0.55/1M in · $1.65/1M out
Write a CUDA kernel for fused top-k sampling.
X5
Here's a fused kernel that keeps the logits in shared memory and reduces in a single pass:
__global__ void fused_topk(float* logits, int k) {
  __shared__ float buf[1024];
  // warp-level bitonic sort, no global sync
  ...
1,204 tokens · $0.0021 · 96 tok/s
+Message ex1-72b…
// EX5 CLI

Your terminal is the console

Everything on ex5.ai is scriptable: rent GPUs, stream logs, launch fine-tunes, deploy endpoints. And ex5 code is an agentic coding session in your repo, powered by EX-Code 32B.

  • ex5 gpu — rent & ssh in seconds
  • ex5 tune — managed fine-tuning
  • ex5 deploy — models, containers, endpoints
  • ex5 code — agentic coding in your repo
ex5 — zsh
$ ex5 login
✔ authenticated as dev@acme.co · balance $48.20

$ ex5 gpu rent --type rtx4090 --count 2 --max-price 0.40
✔ 2× RTX 4090 @ $0.34/hr · tr-izmir-2 · ssh ready in 41s

$ ex5 tune start --base ex1-72b --data ./support.jsonl
⠸ ft-2841 running · epoch 2/3 · loss 0.412

$ ex5 deploy ft-2841 --name acme-support
✔ live → api.ex5.ai/v1 · "model": "acme-support"

$ ex5 code "fix the race condition in worker.rs"
agent reading worker.rs · 3 files changed · tests passing ✔
▊
// OPEN SOURCE

Our research is public

Weights, serving stack and tooling — free to download, self-host and modify. Or skip the setup and deploy any repo to a marketplace GPU in one click.

github.com/ex5-ai →
ex5-ai/ex1SOON

EX-1 serving recipes and eval harness.

Pythonfree forever
$ git clone github.com/ex5-ai/ex1Deploy on GPU →
ex5-ai/ex5-cliSOON

The ex5 command line — gpu, tune, deploy, code.

TypeScriptfree forever
$ git clone github.com/ex5-ai/ex5-cliDeploy on GPU →
ex5-ai/templatesSOON

The template catalog behind one-click deploys.

Shellfree forever
$ git clone github.com/ex5-ai/templatesDeploy on GPU →
// HOW IT WORKS

One balance, three steps

01
Load credits

Card or wire — prepaid credits in seconds, $5 minimum.

02
Pick your layer

Raw GPUs, the model API, a fine-tune job or the chat workspace — mix freely, same balance.

03
Ship & scale

Per-second GPU billing, per-token inference, one invoice. Scale to zero when you're done.

EX5.ai

Start building on ex5.ai

$5 free credits on signup. A 4090 for 15 hours, or about 9M tokens of EX-1 — your call.