--- license: apache-2.0 datasets: - vanta-research/human-ai-collaboration-2 language: - en base_model: - mistralai/Ministral-3-8B-Reasoning-2512 base_model_relation: finetune library_name: peft tags: - text-generation-inference - logical-reasoning - chat - text - conversational-ai - vanta-research - core-reasoning - cognitive-architecture - persona - reasoning - LLM - vanta-research - apollo-astralis - large-language-model - collaborative-ai - conversational - lora - mistral - ministral - apache - roleplay - research pipeline_tag: text-generation ---
![vanta_trimmed](https://cdn-uploads.huggingface.co/production/uploads/686c460ba3fc457ad14ab6f8/hcGtMtCIizEZG_OuCvfac.png)

VANTA Research

Independent AI research lab building safe, resilient language models optimized for human-AI collaboration

Website Join Us Merch X GitHub

--- ## Apollo Astralis 2 Apollo Astralis 2 is a fine-tuned language model built on the new Ministral 3 8B Reasoning architecture, optimized for: - **Logical reasoning and inference** - **Scientific and mathematical problem-solving** - **Commonsense understanding** - **Multi-step analytical thinking** - **Collaborative problem-solving** This model represents a 10% performance improvement over it's previous iteration, with significant gains across reasoning benchmarks while maintaining strong general capabilities. --- ### Model Details - **Model Name**: Apollo Astralis 2 - **Developer**: VANTA Research - **Base Model**: Ministral-3-8B-Reasoning-2512 - **Training Method**: Low-Rank Adaptation (LoRA) - **Parameters**: 8B base + 70.5MB LoRA adapter - **Training Data**: Custom in-house synthetic data generation containing ~26,000 examples across reasoning, logic, math, and science domains ### Dataset Composition - **Logical Reasoning** - **PIQA** - **Mathematics** - **Science & Commonsense** - **CommonsenseQA** - **WinoGrande** - **Human-AI Collaboration** - **Identity & Persona** --- ## Benchmark Results | Benchmark | Apollo Astralis 1 | Apollo Astralis 2 | Δ | |-----------|-------------------|-------------------|---| | **PIQA** | 90.0% | 90.0% | — | | **WinoGrande** | 30.0% | **40.0%** | **+10.0%** | | **CommonsenseQA** | 50.0% | **70.0%** | **+20.0%** | | **Average** | 56.7% | **66.7%** | **+10.0%** | --- ## Quick Start ```python import torch from transformers import AutoTokenizer, BitsAndBytesConfig, Mistral3ForConditionalGeneration # Note: PEFT not needed - this is the full merged model! # Configure 4-bit quantization bnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True, ) # Load model base_model = Mistral3ForConditionalGeneration.from_pretrained( "Ministral-3-8B-Reasoning-2512", quantization_config=bnb_config, device_map="auto", trust_remote_code=True, torch_dtype=torch.float16, ) model = Mistral3ForConditionalGeneration.from_pretrained( "vanta-research/apollo-astralis-2", quantization_config=bnb_config, device_map="auto", trust_remote_code=True, torch_dtype=torch.float16, ) tokenizer = AutoTokenizer.from_pretrained("vanta-research/apollo-astralis-2") model.eval() ``` ## Examples ### Logical Reasoning ```python prompt = """If all roses are flowers, and some flowers fade quickly, can we conclude that some roses fade quickly? Explain your reasoning.""" # Apollo's response includes: # - Clear problem breakdown # - Syllogistic structure analysis # - Identification of logical fallacy # - Final conclusion with explanation ``` ### Mathematical Problem Solving ```python prompt = """A store offers 25% off, then an additional 10% off the sale price. Is this the same as 35% off? Show your work.""" # Apollo's response includes: # - Step-by-step calculation # - Comparison of compound vs simple discounts # - Clear final answer # - Practical explanation of why they differ ``` ### Creative Problem Solving ```python prompt = """I have a 3-liter jug and a 5-liter jug. How can I measure exactly 4 liters?""" # Apollo's response includes: # - Systematic approach # - Step-by-step solution # - Explanation of mathematical principles # - Enthusiastic encouragement ``` --- ### Technical Limitations - **Memory**: Requires ~16GB for full precision inference (less with quantization) - **Speed**: Response generation may be slower due to chain-of-thought reasoning - **Deployment**: Best served via Ollama or HuggingFace; other formats may require conversion ## Ethical Considerations ### Responsible Use - **Educational Focus**: Designed for learning and exploration, not professional advice - **Verification Required**: Always verify critical information, especially in technical domains - **Personality Awareness**: Warm tone should not be mistaken for emotional capacity or consciousness - **Bias Acknowledgment**: May reflect biases from base model and training data ### Intended Use Cases **Appropriate**: - Educational tutoring and homework help - Learning reasoning and problem-solving skills - Brainstorming and collaborative thinking - Prototyping and development assistance - Research into AI reasoning and persona stability **Inappropriate**: - Professional legal, medical, or financial advice - Critical decision-making without human oversight - High-stakes applications without verification - Contexts requiring formal, clinical communication ## Citation ```bibtex @misc{apollo-astralis-2, title={Apollo Astralis 2}, author={VANTA Research}, year={2025}, url={https://huggingface.co/vanta-research/apollo-astralis-2}, } ``` ## License Apache 2.0 --- ## Contact - Organization: hello@vantaresearch.xyz - Engineering/Design: tyler@vantaresearch.xyz *Proudly developed by VANTA Research in Portland, Oregon*