Compare LLM Models
Select 2-5 models to compare side by side.
Select models to compare
0 / 5 selected
Top downloads
Select 2-5 models
| Attribute |
baidu
|
|---|---|
| Company | baidu |
| Release Date | 2025-06-30 |
| Parameters (B) | 0.4 |
| Architecture | Transformer |
| Context Window | 131072 |
| Input Modalities | text |
| Open Source | Yes |
| License | apache-2.0 |
| Score | 0 |
| VRAM (GB) | 2 |
| Compute | Runs on consumer GPUs with ~2 GB VRAM in BF16/FP16; compatible with vLLM, Hugging Face Transformers, and FastDeploy, with optional W8A16/FP8 quantization |
| Benchmark: AGIEval | 28.5 |
| Benchmark: ARC | 40.6 |
| Benchmark: BBH | 30.4 |
| Benchmark: C-Eval | 40.7 |
| Benchmark: DROP | 28.6 |
| Benchmark: GSM8K | 25.2 |
| Benchmark: HellaSwag | 33 |
| Benchmark: HumanEval | 25 |
| Benchmark: MATH | 12.4 |
| Benchmark: MBPP | 40.2 |
| Benchmark: MMLU | 41.9 |
| Benchmark: MMLU-Pro | 16 |
| Benchmark: SimpleQA | 1.8 |
| Pricing (Input) | — |
| Pricing (Output) | — |