# New Simulation-paired Reservoir pre-released!

**URL:** <https://melissa.discourse.group/t/new-simulation-paired-reservoir-pre-released/56>\
**Category:** Deep Learning\
**Created:** [October 21, 2025, 12:15pm UTC](https://melissa.discourse.group/t/new-simulation-paired-reservoir-pre-released/56 "2025-10-21T12:15:18Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![AbhishekP](https://avatars.discourse-cdn.com/v4/letter/a/aeb1de/32.png) [@AbhishekP](https://melissa.discourse.group/u/AbhishekP)\
**Post date:** [October 21, 2025, 12:15pm UTC](https://melissa.discourse.group/t/new-simulation-paired-reservoir-pre-released/56/1 "2025-10-21T12:15:18Z")

</div>

## Main Updates

These updates are related to the new reservoir implementation that accommodates the foundational model training such as:

- [**ScOT**](https://arxiv.org/abs/2405.19101): An All-to-All Architecture with no conditioning other than input time
- [**PDE-Transformer**](https://arxiv.org/abs/2505.24717): An autoregressive Architecture incorporating PDE-type conditioning mechanisms.

### Client-side changes

- Ability to send a JSON-string as metadata alongside the data vectors through `melissa_send_with_metadata()` in C and overloading `melissa_send` for the Python API. ScOT can train several PDEs and thus expects a PDE-type metadata as a label for the data/mesh sent for training.

```python
metadata = {
    "pde": pde_type,
    "num_channels": num_channels,
    "resolution": num_points,
    "simulation_dt": dt,
    "fields": fields,
    "constants": constants,
    "velocity x": vel[0],
    "velocity y": vel[1],
    "seed": seed,
}
metadata_json = json.dumps(metadata)
melissa_send("temperature", data, metadata_json)

```

_Example demonstrates usage of some static values, but preferably send values that change across simulations or its time steps._

### Server-side changes

- Inclusion of `SimPairedReservoir`
- Addition of a new `dataclass` called `Payload`.
- Now, the `SimulationData` contains the `Payload` instance and therefore should be accessed as follows:

```python
data: np.ndarray = msg["temperature"].data
metadata: Dict[str, Any] = msg["temperature"].metadata

```

## Minor Updates

- Data type fixes for the socket types in `melissa_data.h`
- METADATA file `codemeta.json` integration for the general Melissa metadata and information access.