> ## Documentation Index
> Fetch the complete documentation index at: https://bagel.softnanolab.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Output Types

> Shared output types, cross-model field comparison, and include_fields filtering for boileroom models.

All boileroom models return typed dataclass objects containing prediction results and metadata. Each model's output type is documented on its own page — this page covers the shared `PredictionMetadata` type, cross-model comparison, and the `include_fields` filtering mechanism.

## PredictionMetadata

Every output's `metadata` field is a `PredictionMetadata` instance with timing and model information.

<ResponseField name="model_name" type="str">
  Name of the model (e.g., `"ESMFold"`, `"ESM-2"`, `"Chai-1"`, `"Boltz-2"`).
</ResponseField>

<ResponseField name="model_version" type="str">
  Version string of the model.
</ResponseField>

<ResponseField name="sequence_lengths" type="list[int] | None">
  Number of residues for each input sequence (excluding chain separators).
</ResponseField>

<ResponseField name="preprocessing_time" type="float | None">
  Time spent in preprocessing (seconds).
</ResponseField>

<ResponseField name="inference_time" type="float | None">
  Time spent in model inference (seconds).
</ResponseField>

<ResponseField name="postprocessing_time" type="float | None">
  Time spent in postprocessing (seconds).
</ResponseField>

## Per-model output types

Each model page documents its full output dataclass:

* [`ESMFoldOutput`](/boileroom-api/models/esmfold#output) — returned by `ESMFold.fold()`
* [`ESM2Output`](/boileroom-api/models/esm2#output) — returned by `ESM2.embed()`
* [`Chai1Output`](/boileroom-api/models/chai1#output) — returned by `Chai1.fold()`
* [`Boltz2Output`](/boileroom-api/models/boltz2#output) — returned by `Boltz2.fold()`

## Cross-model comparison

| Field           | ESMFold | ESM2 | Chai-1 | Boltz-2 |
| --------------- | ------- | ---- | ------ | ------- |
| `metadata`      | Yes     | Yes  | Yes    | Yes     |
| `atom_array`    | Yes     | —    | Yes    | Yes     |
| `embeddings`    | —       | Yes  | —      | —       |
| `plddt`         | Yes     | —    | Yes    | Yes     |
| `pae`           | Yes     | —    | Yes    | Yes     |
| `pde`           | —       | —    | Yes    | Yes     |
| `ptm`           | Yes     | —    | Yes    | —       |
| `iptm`          | —       | —    | Yes    | —       |
| `confidence`    | —       | —    | —      | Yes     |
| `pdb`           | Yes     | —    | —      | Yes     |
| `cif`           | Yes     | —    | Yes    | Yes     |
| `chain_index`   | Yes     | Yes  | —      | —       |
| `residue_index` | Yes     | Yes  | —      | —       |
| `hidden_states` | —       | Yes  | —      | —       |
| `s_s` / `s_z`   | Yes     | —    | —      | —       |

## Filtering with `include_fields`

All models support an `include_fields` configuration key that controls which optional fields are returned. This is useful for reducing memory usage and transfer time when you only need specific outputs.

```python theme={null}
# Only return pLDDT scores (atom_array is always included for folding models)
result = model.fold("MKTVRQERLKSIVRI", options={"include_fields": ["plddt"]})

# Return everything including PDB/CIF strings
result = model.fold("MKTVRQERLKSIVRI", options={"include_fields": ["*"]})

# Return PDB and PAE
result = model.fold("MKTVRQERLKSIVRI", options={"include_fields": ["pdb", "pae"]})
```

Fields not listed in `include_fields` are set to `None`. The following fields are always included regardless of the filter:

* **Folding models:** `metadata`, `atom_array`
* **Embedding models:** `metadata`, `embeddings`, `chain_index`, `residue_index`