Gravity 16B A3B Base is an open-source language model by trillionlabs. Features: 16b LLM, VRAM: 32.4GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.36.
Arxiv:2412.19437 Arxiv:2502.16982 Base Custom code En Gravity Gravity moe Ko Mixture-of-experts Moe Pretrained Region:us Safetensors Sharded Tensorflow Trillion-labs
Model Card on HF 🤗: https://huggingface.co/trillionlabs/Gravity-16B-A3B-Base
Gravity 16B A3B Base Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Gravity 16B A3B Base Parameters and Internals
| LLM Name | Gravity 16B A3B Base |
| Repository 🤗 | https://huggingface.co/trillionlabs/Gravity-16B-A3B-Base |
| Model Size | 16b |
| Required VRAM | 32.4 GB |
| Updated | 2026-04-10 |
| Maintainer | trillionlabs |
| Model Type | gravity_moe |
| Model Files | |
| Supported Languages | en ko |
| Model Architecture | GravityMoEForCausalLM |
| License | apache-2.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151552 |
| Torch Data Type | bfloat16 |
Best Alternatives to Gravity 16B A3B Base
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| L1 16B A3B | 32K / 32.5 GB | 27 | 5 |
Rank the Gravity 16B A3B Base Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 52721 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a