Space Science & Astronomy

Access NASA, ESA, JWST, Hubble, LIGO gravitational waves, and exoplanet databases — analyze space data on your private GPU cluster.

NASA/ESA JWST Data LIGO Waves Exoplanets Air-Gapped

15+

Space agencies

JWST

Real-time data

LIGO

GW events

Cloud cost

Distributed GPU Cluster

Eldric runs on any GPU — from a single RTX 3090 to NVIDIA H100/H200 datacenter cards. Connect labs, datacenters, and remote researchers into one cluster that spans cities or continents. Workers register through the Edge TLS gateway over the internet, or use the built-in tunnel for NAT traversal — no VPN needed. Adding a node is one command: the worker auto-registers and is immediately available for inference.

# Node 1: Controller + Router + Data Worker (any machine)
./eldric-controller --port 8880 &
./eldric-routerd --controller http://localhost:8880 &
./eldric-datad --nfs --vector --controller http://localhost:8880 &

# Add a GPU worker — any machine, any GPU.
# It auto-registers with the controller and is immediately available.
./eldric-workerd --backend ollama --controller http://node1:8880

# Add more workers the same way. Mix any hardware:
ssh rtx-box   "./eldric-workerd --backend ollama --controller http://node1:8880"
ssh h100-node "./eldric-workerd --backend vllm  --controller http://node1:8880"
ssh mac-studio "./eldric-workerd --backend mlx  --controller http://node1:8880"

# Science Worker (auto-registers like any other worker)
./eldric-scienced --controller http://node1:8880

# Split a 70B model across all workers (VRAM-proportional sharding)
curl -X POST http://node1:8880/api/v1/pipeline/deploy \
  -d '{"model_path":"/mnt/models/llama-70B-Q4.gguf",
       "workers":["wrk-1","wrk-2","wrk-3","wrk-4"],
       "strategy":"vram_proportional"}' 

xLSTM + Transformer Mixture Models

Train xLSTM on photometric time-series for exoplanet transit detection and gravitational wave classification. The long-context sLSTM cells capture periodic signals across months of observation data.

Architecture

Sepp Hochreiter's xLSTM extended with Transformer layers for domain-specific tasks.

sLSTM: exponential gating for long-range dependencies
mLSTM: matrix memory with covariance update
Transformer: cross-attention for multi-feature correlation
Mixture: xLSTM temporal + Transformer relational

Training Configuration

# Train via Eldric Training Worker API
curl -X POST http://controller:8880/api/v1/training/jobs \
  -H "Content-Type: application/json" \
  -d '{"name":"xlstm-exoplanet-transit","base_model":"xlstm-500m","method":"sft","backend":"xlstm","dataset":{"path":"/data/kepler-lightcurves.jsonl","format":"timeseries"},"hyperparams":{"epochs":30,"batch_size":16,"learning_rate":1e-4},"model_config":{"architecture":"xlstm_transformer_mixture","xlstm_layers":8,"transformer_layers":2,"hidden_size":256,"context_length":16384}}'

API Examples

All endpoints are served by the Science Worker (:8897). Requests are routed through the Edge server for TLS and authentication in production.

Astronomy Picture of the Day

GET http://science-worker:8897/api/v1/nasa/apod

Search JWST observations

POST http://science-worker:8897/api/v1/jwst/observations
# Request body:
{"instrument":"NIRCam","target":"TRAPPIST-1","date_range":["2024-01-01","2024-12-31"]}

Search gravitational wave events

POST http://science-worker:8897/api/v1/ligo/search
# Request body:
{"event_type":"BBH","min_mass":20,"confidence":0.9}

SDSS sky survey query

POST http://science-worker:8897/api/v1/sdss/query
# Request body:
{"ra":180.0,"dec":45.0,"radius":0.1,"type":"galaxy"}

Distributed Inference Documentation

Learn how to split large models across multiple workers with pipeline parallelism.

Distributed Inference Docs