Model Card on HF 🤗: https://huggingface.co/deepseek-ai/deepseek-llm-7b-base
Deepseek Llm 7B Base Benchmarks
MMLU Pro: 8.96
GPQA: 3.13
MUSR: 3.76
BBH: 9.77
IFEval: 21.79 vs 88 (so35)-75.2%
ARC: 68.3 vs 96.7 (so35)-29.4%
HellaSwag: 73.4 vs 95.3 (gpt4)-23%
MMLU: 49.3 vs 88.3 (so35)-44.2%
GSM8K: 18.3 vs 96.4 (so35)-81%
OpenCompas: 44.4 vs 73.3 (gpt4)-39.4%
Exam: 47.3 vs 77.2 (gpt4)-38.7%
Language: 37.4 vs 62 (gpt4)-39.7%
Knowledge: 46.5 vs 73.5 (gpt4)-36.7%
Understand: 51 vs 70 (gpt4)-27.1%
Reasoning: 42.5 vs 74.4 (gpt4)-42.9%
MATH Lvl 5: 1.96
LLME Score: 0.19959
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)
Deepseek Llm 7B Base Parameters and Internals
| Model Type |
| |||||||||
| Use Cases |
| |||||||||
| Supported Languages |
| |||||||||
| Training Details |
|
| LLM Name | Deepseek Llm 7B Base |
| Repository 🤗 | https://huggingface.co/deepseek-ai/deepseek-llm-7b-base |
| Model Size | 7b |
| Required VRAM | 13.9 GB |
| Updated | 2025-09-23 |
| Maintainer | deepseek-ai |
| Model Type | llama |
| Model Files | |
| Model Architecture | LlamaForCausalLM |
| License | other |
| Context Length | 4096 |
| Model Max Length | 4096 |
| Transformers Version | 4.33.1 |
| Tokenizer Class | LlamaTokenizerFast |
| Beginning of Sentence Token | <｜begin▁of▁sentence｜> |
| End of Sentence Token | <｜end▁of▁sentence｜> |
| Vocabulary Size | 102400 |
| Torch Data Type | bfloat16 |
Quantized Models of the Deepseek Llm 7B Base
Model |
Likes |
Downloads |
VRAM |
|---|---|---|---|
| Deepseek Llm 7B Base GGUF | 6 | 1445 | 3 GB |
| Deepseek Llm 7B Base GPTQ | 2 | 1072 | 4 GB |
| Deepseek Llm 7B Base AWQ | 2 | 138 | 4 GB |
Best Alternatives to Deepseek Llm 7B Base
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| A6 L | 1024K / 16.1 GB | 201 | 0 |
| A3.4 | 1024K / 16.1 GB | 13 | 0 |
| A5.4 | 1024K / 16.1 GB | 12 | 0 |
| A2.4 | 1024K / 16.1 GB | 12 | 0 |
| M | 1024K / 16.1 GB | 127 | 0 |
| 157 | 1024K / 16.1 GB | 101 | 0 |
| 124 | 1024K / 16.1 GB | 93 | 0 |
| 162 | 1024K / 16.1 GB | 60 | 0 |
| 2 Very Sci Fi | 1024K / 16.1 GB | 317 | 0 |
| 118 | 1024K / 16.1 GB | 15 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than deepseek-ai/deepseek-llm-7b-base .
Rank the Deepseek Llm 7B 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? 51534 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124