> ## Documentation Index > Fetch the complete documentation index at: https://docs.platform.toloka.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # How Training works > Understand Training methods, data, lifecycle, run states, outputs, and supported product surfaces. Training trains a model on a frozen slice of a dataset in your project. You start from a catalog Base model or from one of your own LoRA adapter Model Versions. A free cost estimate validates the inputs. A billed Training Run publishes a Model Version when it completes. The output is a set of model files that you download and deploy yourself. The platform does not host or serve it. ## Two methods | Method | What it reads | What it produces | | - | - | - | | Supervised fine-tuning (SFT) | Conversations in a dataset | A LoRA adapter that runs on top of the Base model | | Prompt gisting | Conversations in a dataset plus a prompt text file | A full model containing learned gist tokens | SFT is the default method in the CLI. Gisting supports only the `qwen3-4b` and `qwen3-8b` Base models because it publishes full weights. Both methods can start from a LoRA adapter Model Version in the same project. See [Choose a training method](https://docs.platform.toloka.ai/docs/train/get-started/choose-method.md) for a decision guide. ## From draft to Model Version A **Training** is a project-owned, editable draft. It holds the method, model source, dataset slice, split column, and settings. Its name is 1–120 characters and is unique among active Trainings in the project. You can archive or rename a Training through the HTTP API. Editing a Training does not change an existing Training Run. A **cost estimate** prepares the selected rows from the frozen slice and validates them. It is either blocked, with issues to fix, or ready, with a maximum charge. * A ready estimate expires 24 hours after you request it. * A Training can have at most three open estimates. An estimate remains open until it starts a Training Run or expires. * One ready estimate starts one Training Run. * Editing the Training makes an earlier estimate unusable. Calculate it again. A **Training Run** is one immutable execution. It copies the Training configuration when created. Starting it places a hold for the maximum charge on the team balance. See [Training cost](https://docs.platform.toloka.ai/docs/train/concepts/cost.md) for estimation, holds, settlement, and no-charge cases. The first Training Run creates a **Model** named after the Training. Each completed Training Run publishes one immutable **Model Version** in that Model. Revisions are numbered 1, 2, 3, and so on. Model Version names must be unique within the Model. In the CLI, the default version name is the Training name; pass `--version-name` after an earlier run has used it. ## Data A dataset row is one chat example. This row teaches the model to label a support ticket: ```json theme={null} { "messages": [ { "role": "user", "content": "Hi, how do I reset my password? I can't find the option anywhere." }, { "role": "assistant", "content": "question" } ] } ``` Each selected row must follow these rules: * The row has `messages` and can also have `tools`. * Allowed roles are `system`, `user`, `assistant`, and `tool`. * At least one turn has the `user` role. * The last turn has the `assistant` role and contains content. * System turns appear only before the conversation starts. * Consecutive turns cannot have the same role. Tool and system turns are exempt. * Every `system`, `user`, and `tool` turn has content. An intermediate `assistant` turn has content or tool calls. * Each `tool` turn answers a preceding assistant tool call. * Every assistant tool call is answered by a later `tool` turn. A dataset can come from uploaded JSONL rows that contain `messages`. It can also keep a conversation in a `messages` column and the expected answer in a label column. The platform appends that label as the final assistant turn. The CLI can create an empty dataset and append JSONL rows to it: ```sh theme={null} toloka dataset create support-tickets toloka dataset upload support-tickets ./support-tickets.jsonl ``` Uploads append. Running the upload twice adds the rows twice. Both methods read one ready, frozen dataset slice that includes a split column. The split assigns rows to named parts with these roles: | Role | Use in Training | | - | - | | `train` | Rows the model learns from. By default, Training uses every part with this role. | | `validate` | Rows used for validation metrics. Training uses all parts with this role. | | `test` | Rows kept out of learning. | Pass `--dataset ` to prepare the dataset's split column and freeze a slice named after the Training. If the dataset has no split column, the CLI creates an 80% train, 10% validation, 10% test split. Existing assignments stay in place. Pass `--slice ` instead to reuse a ready slice. Use `--split-column` to choose among multiple split columns and `--parts` to select particular train parts; see [Dataset selections](https://docs.platform.toloka.ai/docs/train/reference/settings.md#dataset-selections). To set your own shares or freeze part of a dataset, see [Slices and split columns](https://docs.platform.toloka.ai/docs/train/reference/settings.md#slices-and-split-columns). The slice created by `--dataset` fixes the rows and field values the Training reads. Later uploads or label changes are not included automatically. Run `toloka training update --dataset ` to freeze a new slice, then calculate a new estimate. If the slice supplies no validation rows, the platform holds out 10% of the selected training rows for validation. The cost estimate checks row shape, token length, selection size, batch size, export success, and the estimated time limit. See [Training limits](https://docs.platform.toloka.ai/docs/train/reference/limits.md) and [Training errors](https://docs.platform.toloka.ai/docs/train/reference/errors.md). For the path from labeled project data to valid training examples, see [Prepare training data](https://docs.platform.toloka.ai/docs/train/guides/prepare-training-data.md). ## Training Run states Use `toloka training status ` to check progress, `toloka training logs ` for sanitized logs, and `toloka training metrics ` for reported metrics. A Training Run normally moves from `queued` to `running` to `completed`. Other states represent a stop or a billing hold. | State | Meaning | Action | | - | - | - | | `queued` | Preparing data and reserving the maximum charge. | Wait, or cancel without charge. | | `running` | Waiting for GPUs, training, checking output, settling the charge, or publishing. | Monitor progress. Cancel is available only before output checks begin. | | `blocked`, reason `funds` | The balance cannot cover the maximum charge. | Add funds and resume, or cancel. | | `blocked`, reason `payment` | The final charge could not be collected. | Add funds and resume before the payment window ends. Cancel is unavailable. | | `blocked`, reason `billing_review` | Billing needs review by Toloka staff. | Wait for resolution. Resume and cancel are unavailable. | | `stopping` | The Training Run is stopping after cancellation or failure and settling billing. | Wait. | | `completed` | The Model Version is published. | Inspect or download it. | | `failed` | The Training Run ended with an error. | Inspect the error. It cannot resume. | | `canceled` | The Training Run was canceled. | It cannot resume. | While status is `running`, the `Progress` row can show `preparing`, `queued`, `pending`, `training`, `evaluating`, or `publishing`. Progress `queued` or `pending` means the job is waiting for GPU capacity. That wait can take hours, and the job starts automatically when capacity is available. Metrics stop updating when training ends. Output checks, billing, and publication still need to finish before status becomes `completed`. If metrics report `not_configured`, the Training Run will not produce metrics. If they report `unresolved`, check again while the Training Run is running. For a failed run, inspect `toloka training logs ` and look up the reported error in [Training errors](https://docs.platform.toloka.ai/docs/train/reference/errors.md). ## Cancellation and payment holds Canceling before training starts has no charge. Canceling during training stops the job, charges for billable GPU use up to the approved maximum, and publishes no Model Version. A Training Run canceled while it reserves funds or trains shows `stopping` until the job stops and billing settles; canceled while preparing data or blocked on funds, it becomes `canceled` at once. Cancellation is no longer available once output validation starts or while final payment is blocked. The payment window is seven days. If it expires, the Training Run fails with `payment_window_expired`, cannot resume, and is not charged. Its billing status remains `payment_required`. ## Outputs and downloads An SFT Model Version includes `adapter_model.safetensors` and `adapter_config.json`. It can contain additional `.json` files. It is a LoRA adapter and needs the matching Base model. The Model Version records the Base model ID; the catalog carries the current upstream repository and revision. The adapter configuration may omit the revision. See [Pinned revisions and lineage](https://docs.platform.toloka.ai/docs/train/reference/base-models.md#pinned-revisions-and-lineage) before serving it. A gisting Model Version includes `config.json` and either `model.safetensors` or a complete numbered set of `model-*.safetensors` weight shards. Tokenizer files, `merges.txt`, and `chat_template.jinja` are optional. Download a completed Model Version with: ```sh theme={null} toloka model version download --out ``` The CLI requests fresh download links and verifies each file's size and SHA-256. If a link expires, rerun the download command. ## Ways to work with Training Project Training is available through: * [CLI](https://docs.platform.toloka.ai/docs/cli.md) — create and update Trainings, calculate cost, start and control Training Runs, inspect logs and metrics, and download Model Versions * [Fine-tuning agent runbook](https://platform.toloka.ai/agents/fine-tuning.md) — guide a coding agent through an SFT workflow * [Gisting agent runbook](https://platform.toloka.ai/agents/gisting.md) — guide a coding agent through a gisting workflow * [HTTP API](https://docs.platform.toloka.ai/docs/api-reference.md) — work with Training, Training Run, Base model, Model, and Model Version resources In the web app, manage project datasets and team balance. Open **Fine-tuning** or **Prompt gisting** to set up a coding agent. If you see **Fine-tune lab**, it offers a separate web workflow with its own runs and downloads. Use the CLI or HTTP API for project Training. ## Related * [Choose a training method](https://docs.platform.toloka.ai/docs/train/get-started/choose-method.md) * [Training cost](https://docs.platform.toloka.ai/docs/train/concepts/cost.md) * [Base models](https://docs.platform.toloka.ai/docs/train/reference/base-models.md) * [Training settings](https://docs.platform.toloka.ai/docs/train/reference/settings.md) * [Training limits](https://docs.platform.toloka.ai/docs/train/reference/limits.md) * [Train your first model](https://docs.platform.toloka.ai/docs/train/get-started/first-trained-model.md) * [Label, train, and evaluate](https://docs.platform.toloka.ai/docs/workflows/label-train-evaluate.md) This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.