GPT-2 models
8 GPT-2 models from openai-community on Hugging Face, from 110M to 1.6B parameters, published by openai-community in F32, with a 1K-token context. The smallest official model, gpt2, needs about 0.6 GB of VRAM at its published precision; the cheapest live fit is RTX 3070 at $0.088/hr.
Pick a size
One row per official GPT-2 size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| gpt2 | 137M | 0.6 GB | 0.2 GB | 0.1 GB | RTX 3070 | $0.088/hr | 1.1 GB |
| gpt2-medium | 380M | 1.7 GB | 0.4 GB | 0.2 GB | RTX 3070 | $0.088/hr | 3.0 GB |
| gpt2-large | 812M | 3.6 GB | 0.9 GB | 0.5 GB | RTX 3070 | $0.088/hr | 5.6 GB |
| gpt2-xl | 1.6B | 7.2 GB | 1.8 GB | 0.9 GB | RTX 3070 | $0.088/hr | 9.4 GB |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (4)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| gpt2 | 137M | F32 | 0.6 GB | RTX 3070 | $0.088/hr |
| gpt2-medium | 380M | F32 | 1.7 GB | RTX 3070 | $0.088/hr |
| gpt2-large | 812M | F32 | 3.6 GB | RTX 3070 | $0.088/hr |
| gpt2-xl | 1.6B | F32 | 7.2 GB | RTX 3070 | $0.088/hr |
Fine-tunes and community models (4)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| vietnamese-gpt2 | 110M | F32 | 0.5 GB | RTX 3070 | $0.088/hr |
| turkish-gpt2 | 124M | F32 | 0.6 GB | RTX 3070 | $0.088/hr |
| gpt2-small-dutch | 129M | F32 | 0.6 GB | RTX 3070 | $0.088/hr |
| megatron-gpt2-345m | 380M | F16 | 0.8 GB | RTX 3070 | $0.088/hr |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
More from openai-community
- openai-community (1 models)