Autotrain compatible Conversational En Endpoints compatible Llama2 Mistral Region:us Safetensors Sharded Starling Tensorflow
Model Card on HF 🤗: https://huggingface.co/Delcos/Velara
Velara Benchmarks
ARC: 58.96 vs 96.7 (so35)-39%
HellaSwag: 82.83 vs 95.3 (gpt4)-13.1%
MMLU: 59.45 vs 88.3 (so35)-32.7%
TruthfulQA: 44.7 vs 59 (gpt4)-24.2%
WinoGrande: 73.8 vs 87.5 (gpt4)-15.7%
GSM8K: 40.33 vs 96.4 (so35)-58.2%
LLME Score: 0.1546
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)
Velara Parameters and Internals
| Model Type |
| |
| Additional Notes |
| |
| Supported Languages |
|
| LLM Name | Velara |
| Repository 🤗 | https://huggingface.co/Delcos/Velara |
| Model Size | 11.4b |
| Required VRAM | 22.8 GB |
| Updated | 2025-09-21 |
| Maintainer | Delcos |
| Model Type | mistral |
| Model Files | |
| Supported Languages | en |
| Model Architecture | MistralForCausalLM |
| License | cc-by-nc-nd-4.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.35.2 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <|im_end|> |
| Vocabulary Size | 32002 |
| Torch Data Type | bfloat16 |
Quantized Models of the Velara
Model |
Likes |
Downloads |
VRAM |
|---|---|---|---|
| Velara GGUF | 4 | 115 | 4 GB |
| Velara GPTQ | 1 | 6 | 6 GB |
| Velara AWQ | 0 | 6 | 6 GB |
Rank the Velara 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? 51507 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124