Latest News

• [2026/11] Inkling 2 Small from Thinking Machines Lab Is Now Available in Snowflake Cortex Training
• [2026/11] Cortex Training Public Preview blog

---

Cortex Training, Snowflake's serverless offering for customizing and deploying open-weight LLMs. With Cortex Training, teams can use familiar training frameworks, or build advanced custom training loops with low-level serverless APIs - all without managing the complexities of GPU provisioning, scheduling, checkpointing and distributed execution.

Cortex Training is built on top of the highly customizable and optimized open-source Arctic Platform backend to bring the flexibility and efficiency of complex, self-managed bare-metal training cluster to a simplified serverless experience. AI teams no longer have to trade simplicity for flexibility or cost efficiency when customizing and deploying models. See our blog for details.

This repository contains the Python SDK, CLI, model catalog, and recipes for using Cortex Training.

What's in this repository

Area	What it contains	Start here
src/cortex_training/	Python SDK, ct CLI, and terminal UI	Python SDK or CLI reference
recipes/	Runnable training loops, framework integrations, evaluations, inference, and deployment workflows	Recipe index
model-catalog/	Job configurations organized by model, context length, GPU type, and training mode	Model catalog
docs/	Getting-started guides, concepts, operational guides, and API references	Documentation index
tests/	SDK, CLI, model-catalog, and recipe tests	Development

Getting Started

Setup

1. Clone and install — this gives you the SDK, CLI, and all recipes:

   git clone https://github.com/snowflakedb/cortex-training.git
   cd cortex-training
   uv venv
   source .venv/bin/activate
   uv pip install -e .

   Alternatively, uv pip install cortex-training installs the SDK and CLI without the recipes.

2. Get access — you need a Snowflake account with Cortex Training enabled. See account setup for the full walkthrough and authentication for PAT generation and connection configuration.

3. Connect — Cortex Training signs in with a standard Snowflake connection profile, the same kind the Snowflake CLI and Python Connector use. The easiest way to set one up is to run:

   cortex-training login

   It creates or picks a profile, checks that it connects to Cortex Training, and saves it as the one to use. Every command, the TUI, and the recipes then use it with no extra arguments:

   cortex-training capacity
   cortex-training status      # which profile, account and database commands use, and why

   To write the profile yourself instead, add it to connections.toml in your Snowflake folder, usually ~/.snowflake/connections.toml. cortex-training login prints the exact file it uses; if it printed another path, use that one:

   [cortex-training]
   account = "ACCOUNT"
   host = "ACCOUNT.snowflakecomputing.com"
   user = "USER"
   role = "ROLE"
   authenticator = "PROGRAMMATIC_ACCESS_TOKEN"
   token = "YOUR_PROGRAMMATIC_ACCESS_TOKEN"
   database = "CORTEX_TRAINING_DB"
   schema = "PUBLIC"

   Then protect the file (use the path login printed, if different) and check the profile:

   chmod 600 ~/.snowflake/connections.toml
   cortex-training login cortex-training

   Pass --connection NAME (or connection=NAME to a recipe) to use another profile for one command. See authentication for other sign-in methods, such as key pairs.

Already using a JSON connection file?

If you connect with a JSON file (host, pat, database, schema), for example through --config, CORTEX_TRAINING_CONFIG, cortex-training login FILE, or a recipe's config=, convert it into a profile with one command:

cortex-training login --convert-legacy-config ~/cortex-training-config.json

It shows what it will change and asks first: it checks the token, adds a profile, saves it as the one to use, and signs in with it. Your JSON file is kept as it was, so you can go back to it with --config FILE. JSON files still work, and will be deprecated in a future release.

Hello World with Cortex Training

Fine-tune a chat model on a small bundled dataset:

uv pip install 'tinker-cookbook @ git+https://github.com/thinking-machines-lab/tinker-cookbook.git@nightly'
python -m recipes.sft_quickstart.conversational.train max_steps=50

It connects with the profile from step 3 and fine-tunes Qwen3-8B (full-parameter, four H200 GPUs) on a small bundled dataset. See the SFT recipes for other datasets, models, and training options.

Choose a recipe type

If you want to…	Recipe type	Start here
Fine-tune on conversations, labeled data, or agent traces	Supervised fine-tuning	SFT recipes
Optimize model behavior using rewards	Reinforcement learning	RL recipes
Transfer behavior from a teacher to a student	Distillation	Distillation recipes
Train a draft model for faster inference	Speculative decoder training	DFlash2 recipe
Serve checkpoints, generate responses, or evaluate models	Inference and evaluation	Inference recipes
Create an OpenAI-compatible endpoint	Model deployment	Deployment recipes
Use SkyRL, Tinker, OpenHands, or another system	Frameworks and agents	Framework recipes
Learn the API building blocks or construct a custom loop	API requests	API recipes

The complete recipe index lists the individual workflows and their entry points.

How recipes work

Recipes are runnable, end-to-end workflows that you can use as written or adapt. They build jobs from the model catalog, which contains configurations organized by model, context length, GPU type, and training mode. Select those options on the command line instead of writing a complete job body.

Each recipe README covers its prerequisites, commands, options, and expected results. See model compatibility for the currently supported combinations.

Build custom workflows

Interface	Best for	Reference
Complete recipes	Starting with an end-to-end workflow and adapting its task or training loop	Recipe index
CortexTrainingClient	Cortex-specific recipes that create and manage job bodies directly, address operations by job_id, or need APIs such as capacity, logs, checkpoints, and model deployment	Python SDK reference
ArcticRLClient	RL recipes that want a model-oriented interface over training, sampling, and log-probability workers, with fwd_bwd, step, generate, log_probs, and sync_weights operations	Arctic Platform RL · Framework recipes
CLI and REST API	Controlling individual job and model operations	CLI reference · REST API reference

Recipes can be built on either client. With CortexTrainingClient, the recipe creates the Cortex Training job and passes its job_id to each operation. With ArcticRLClient, the recipe describes the model and its training, sampling, and optional log-probability workers; the client presents those workers through one RL API. Its backend can be Cortex Training or an on-prem Arctic Platform deployment, so it is also the better fit for portable RL recipes and framework integrations.

Documentation

• New users: account setup and first job
• Coding agents: cortex-training skills install gives your agent a setup skill to check and fix your Cortex Training connection
• Recipe users: complete recipe index
• Model selection: model compatibility and model catalog
• Core concepts: jobs, lifecycle, checkpoints, hardware, and sizing
• Operations: job management, logs and metrics, and troubleshooting
• API development: CLI, Python SDK, and REST API

Development

• Contributing to the client or recipes — see the contributing guides and recipe template. Install development dependencies with uv pip install -e ".[dev]" and run python3 -m pytest tests/ -q.
• Building a distributable package — use ./build_wheel.sh.