Llama 33B Instructed is an open-source language model by Secbone. Features: 33b LLM, VRAM: 65.2GB, Context: 2K, License: gpl-3.0, Quantized, HF Score: 58.2, LLM Explorer Score: 0.14, Arc: 64.6, HellaSwag: 86.2, MMLU: 60.5, TruthfulQA: 44.1, WinoGrande: 79.3, GSM8K: 14.4.
| Model Type |
|
| LLM Name | Llama 33B Instructed |
| Repository ๐ค | https://huggingface.co/Secbone/llama-33B-instructed |
| Model Size | 33b |
| Required VRAM | 65.2 GB |
| Updated | 2026-04-10 |
| Maintainer | Secbone |
| Model Type | llama |
| Model Files | |
| GGML Quantization | Yes |
| GGUF Quantization | Yes |
| Quantization Type | fp16|ggml|q5|q5_1|gguf|f16|q8_0 |
| Model Architecture | LlamaForCausalLM |
| License | gpl-3.0 |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.33.2 |
| Tokenizer Class | LlamaTokenizer |
| Beginning of Sentence Token | <s> |
| End of Sentence Token | </s> |
| Unk Token | <unk> |
| Vocabulary Size | 32000 |
| Torch Data Type | bfloat16 |
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Vicuna 33B V1.3 GGUF | 2K / 34.6 GB | 8 | 0 |
| Guanaco 33B 4bit | 2K / 16.9 GB | 10 | 10 |
| ...Interpreter DS 33B Hf 4bit Mlx | 16K / 19.1 GB | 180 | 7 |
| ...erpreter DS 33B 4.0bpw H6 EXL2 | 16K / 17.1 GB | 3 | 4 |
| ...rpreter DS 33B 4.65bpw H6 EXL2 | 16K / 19.8 GB | 1 | 3 |
| ...erpreter DS 33B 6.0bpw H6 EXL2 | 16K / 25.3 GB | 7 | 1 |
| ...erpreter DS 33B 5.0bpw H6 EXL2 | 16K / 21.2 GB | 1 | 1 |
| ...erpreter DS 33B 8.0bpw H8 EXL2 | 16K / 33.5 GB | 0 | 1 |
| ...se DeepSeek 33B 8.0bpw H8 EXL2 | 16K / 33.5 GB | 8 | 2 |
| ...se DeepSeek 33B 4.0bpw H6 EXL2 | 16K / 17.1 GB | 2 | 2 |
๐ 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! ๐