Models Enterprise Developers About Contact Get Started

NAPH-7B

7 Billion Parameters

Edge Deployment

NAPH-7B is our most compact model, designed specifically for resource-constrained environments where latency and memory efficiency take priority. Despite its smaller footprint, it maintains strong performance on core language tasks through careful architecture optimization and high-quality training data curation.

The model runs comfortably on consumer-grade hardware including gaming laptops with 8GB VRAM when using 4-bit quantization. This makes it ideal for on-device applications, offline processing, and scenarios where internet connectivity is unavailable or undesirable.

Primary Use Cases

On-device inference Mobile applications Edge computing Offline processing Low-latency chat Text summarization

Technical Specifications

Parameters 7B
Context Length 32,768 tokens
Architecture Dense Transformer
Hidden Dimension 4,096
Attention Heads 32
Layers 32
Min VRAM (FP16) 14GB
Min VRAM (4-bit) 6GB

NAPH-70B

70 Billion Parameters

General Purpose

NAPH-70B represents the optimal balance between capability and practical deployment. It delivers performance competitive with leading closed-source models while remaining deployable on standard enterprise GPU infrastructure. This is our most widely deployed model.

The extended 128K context window enables processing of entire codebases, lengthy legal documents, and complex multi-turn conversations without truncation. Native tool use support makes it suitable for agentic applications requiring structured function calling.

Primary Use Cases

Production workloads RAG applications Document analysis Tool use / Agents Multi-turn chat Content generation

Technical Specifications

Parameters 70B
Context Length 131,072 tokens
Architecture Dense Transformer
Hidden Dimension 8,192
Attention Heads 64
Layers 80
Min VRAM (FP16) 140GB
Min VRAM (8-bit) 70GB

NAPH-405B

405 Billion Parameters

Flagship

NAPH-405B is our frontier model, designed to compete with the most capable systems from any AI lab. It achieves state-of-the-art performance across reasoning benchmarks, complex analysis tasks, and creative generation while maintaining full transparency into its architecture and training.

The model excels at tasks requiring deep reasoning, nuanced understanding, and sophisticated output. It's particularly strong on mathematical reasoning, scientific analysis, and complex instruction following. Multi-modal capabilities support image understanding alongside text.

Primary Use Cases

Complex reasoning Scientific research Mathematical analysis Multi-modal tasks Agentic workflows Frontier applications

Technical Specifications

Parameters 405B
Context Length 131,072 tokens
Architecture Dense Transformer
Hidden Dimension 16,384
Attention Heads 128
Layers 126
Min VRAM (FP16) 810GB
Min VRAM (FP8) 405GB

NAPH-Coder

34 Billion Parameters

Code Specialized

NAPH-Coder is purpose-built for software development tasks. Rather than training a general model and hoping it performs well on code, we designed the architecture, curated the training data, and optimized the training process specifically for programming applications.

The model understands 80+ programming languages with deep knowledge of idioms, patterns, and best practices. It excels at code completion, generation, review, refactoring, documentation, and debugging. Extended context allows processing entire codebases for architectural understanding.

Primary Use Cases

Code completion Code generation Code review Refactoring Documentation Bug detection Test generation

Technical Specifications

Parameters 34B
Context Length 65,536 tokens
Architecture Dense Transformer
Hidden Dimension 6,656
Attention Heads 52
Layers 60
Min VRAM (FP16) 68GB
Languages 80+

Performance

Benchmark Results

Performance across standard evaluation benchmarks. All results are from internal evaluations using consistent methodology. We publish full evaluation details in our technical reports.

Benchmark NAPH-7B NAPH-70B NAPH-405B NAPH-Coder
MMLU (5-shot) 63.2% 79.8% 86.4% 71.3%
HumanEval (pass@1) 42.1% 67.4% 81.2% 89.6%
GSM8K (CoT) 58.7% 86.3% 94.1% 78.9%
MATH (4-shot) 24.6% 48.2% 61.8% 43.5%
HellaSwag 79.4% 87.2% 91.3% 84.1%
MBPP (pass@1) 48.3% 72.1% 84.7% 91.2%
WinoGrande 74.8% 83.6% 88.4% 79.2%

Ready to Deploy?

Get API access or talk to our team about on-premises deployment options.