Models / TimeLens2
TimeLens2 by MCG-NJU
6 models (2B to 8B) · ↓ 4.4K downloads a month · from $0.09/h on RTX 3060 right now. Each model's versions do the same job; quantized ones (AWQ, FP8, 4-bit) need less GPU memory and run on cheaper machines.
TimeLens2 2B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| OriginalMCG-NJU/TimeLens2-2B | vLLM | 4.9 GB | 9.7 GB | $0.09/h RTX 3060 |
TimeLens2 2B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| OriginalMCG-NJU/TimeLens2-2B-SFT | vLLM | 4.3 GB | 9 GB | $0.09/h RTX 3060 |
TimeLens2 4B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| OriginalMCG-NJU/TimeLens2-4B | vLLM | 9.7 GB | 15.6 GB | $0.13/h RTX A4000 |
TimeLens2 4B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| OriginalMCG-NJU/TimeLens2-4B-SFT | vLLM | 8.9 GB | 14.8 GB | $0.13/h RTX A4000 |
TimeLens2 8B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| OriginalMCG-NJU/TimeLens2-8B | vLLM | 18 GB | 24.4 GB | $0.27/h 2x RTX 4060 Ti |
TimeLens2 8B
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| OriginalMCG-NJU/TimeLens2-8B-SFT | vLLM | 18 GB | 24.4 GB | $0.27/h 2x RTX 4060 Ti |
Model details come from Hugging Face and are refreshed daily; prices follow the live offers. GPU memory is GridPup's estimate: the weights, room for an 8,192-token conversation (chat models) and the runtime's own memory. Large models are split across the GPUs of one machine ("4x RTX 4090").