esa-sceva/llama3-satcom-70b

LLama3 SatCom 70B

LLama3 SatCom 70B is a high-capacity fine-tuned Large Language Model (LLM) developed under the ESA ARTES programme as part of the SatcomLLM / SCEVA (SatCom Expert Virtual Assistant) project.
It represents the largest and most capable SatCom-specialised LLM, purpose-built to support satellite communications (SatCom) professionals through advanced reasoning, multi-step problem solving, and expert-level Q&A.

---

Model Description

• Base model: meta-llama/Llama-3.3-70B-Instruct
• Fine-tuning type: Instruction fine-tuning (IFT)
• Training data: Curated and synthetic SatCom QA, including chain-of-thought annotated datasets
• Architecture: Decoder-only transformer, 70 billion parameters
• Languages: English
• License: LLama-3.3 Community License Agreement

This large-scale variant extends the reasoning and factual accuracy of the 8B model, providing enhanced comprehension of technical systems, higher stability in multi-step mathematical reasoning, and greater contextual understanding across complex SatCom documents.
It excels in tasks such as link budget evaluation, propagation modeling, 5G/6G NTN design, and mission architecture analysis. Thanks to its larger parameter count and extended context window, it can process longer technical passages, integrate multiple data sources, and maintain coherence across complex analytical workflows.

---

Training Datasets

Dataset	Description
esa-sceva/satcom-synth-qa	Synthetic QA data generated with domain-validated prompts and expert-reviewed teacher models
esa-sceva/satcom-synth-qa-cot	Chain-of-thought annotated QA to strengthen reasoning accuracy and factual traceability

---

Intended Use

Primary use cases:

• Advanced reasoning and Q&A for SatCom system design and analysis
• Automated support for link budget calculations and RF engineering tasks
• Conceptual guidance for 5G/6G NTN and inter-satellite network operations
• Mission design evaluation and anomaly diagnosis assistance
• Research, education, and technical documentation in satellite communications

Intended users:

• ESA engineers and mission planners
• SatCom and aerospace researchers
• System architects and technical operators
• Academic institutions and educational users

---

Limitations

• The model does not access live telemetry or proprietary ESA mission data.
• Generated answers should be validated by domain experts before operational use.
• It is not suitable for safety-critical or real-time decision-making.

---

Technical Details

Parameter	Value
Base Model	Llama 3.3 70B Instruct
Parameters	70 billion
Context length	128k tokens
Precision	bfloat16 / fp16
Framework	Lit-GPT (Lightning AI)
Training infra	EuroHPC MareNostrum5 + AWS EC2
Optimisation	LoRA fine-tuning, cosine LR schedule

---

Evaluation

Evaluation Datasets

The model was evaluated across SatCom-specific and general-domain benchmarks, focusing on mathematical reasoning, conceptual understanding, and applied problem solving.

Dataset	Subset	Description
esa-sceva/satcom-qa	Open SatCom QA	Conceptual and reasoning-based questions on SatCom workflows, regulation, and mission/system design
	Math SatCom QA	Quantitative and multi-step engineering problems derived from orbital mechanics and RF analysis
esa-sceva/satcom-mcqa	Open MCQA	Conceptual multiple-choice questions on protocols, architectures, and communication systems
	Math MCQA	Numerical link-budget and propagation-focused multiple-choice problems

---

Results

Evaluation Subset	Metric	Base Model	LLama3 SatCom 70B	Notes
Math SatCom MCQA	Accuracy	0.664	0.695	Strong quantitative reasoning in link budgets and RF propagation tasks
Open SatCom MCQA	Accuracy	0.944	0.947	High factual precision in conceptual multiple-choice questions
Math SatCom QA	LLM-as-a-judge	0.944	0.955	Excellent performance on expert-reviewed mathematical reasoning tasks
Open SatCom QA	LLM-as-a-judge	0.901	0.911	Robust understanding of terminology, architecture, and system design reasoning

The 70B model delivers improvements in structured reasoning, numerical consistency, and contextual accuracy, making it highly suited for advanced SatCom applications requiring technical depth and reliability.

---

Summary

LLama3 SatCom 70B combines the linguistic precision of Llama 3.3 with domain-specialised fine-tuning from ESA’s SCEVA project, achieving great reasoning and analytical performance in satellite communications.
It represents a step toward intelligent, domain-grounded AI assistants capable of supporting complex engineering and research workflows across the space communications ecosystem.

---