OLMo 1B Hf is an open-source language model by allenai. Features: 1b LLM, VRAM: 4.7GB, Context: 2K, License: apache-2.0, LLM Explorer Score: 0.24, Arc: 34.6, HellaSwag: 63.6, MMLU: 26.3, GSM8K: 1.9.
OLMo 1B Hf Parameters and Internals
| Model Type | | Autoregressive Language Model, Transformer |
|
| Use Cases |
| Areas: | | Research, Commercial Applications |
|
| Applications: | |
| Primary Use Cases: | |
| Limitations: | | Prone to generate false or harmful content |
|
| Considerations: | | Consider risks of applications due to potential bias |
|
|
| Additional Notes | | The model supports quantization for faster inference. |
|
| Supported Languages | |
| Training Details |
| Data Sources: | |
| Data Volume: | | Trained on 3 trillion tokens |
|
| Context Length: | |
| Model Architecture: | |
|
| Input Output | |
| LLM Name | OLMo 1B Hf |
| Repository 🤗 | https://huggingface.co/allenai/OLMo-1B-hf |
| Model Size | 1b |
| Required VRAM | 4.7 GB |
| Updated | 2026-07-27 |
| Maintainer | allenai |
| Model Type | olmo |
| Model Files | 4.7 GB |
| Supported Languages | en |
| Model Architecture | OlmoForCausalLM |
| License | apache-2.0 |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.40.0 |
| Tokenizer Class | GPTNeoXTokenizer |
| Padding Token | <|padding|> |
| Vocabulary Size | 50304 |
| Torch Data Type | float32 |
Best Alternatives to OLMo 1B Hf