nlp-lab5-8-gpt2 is an open-weight model from Fushi. It has 124M parameters. At 16-bit it needs about 0.3 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
Runs On
What it takes to serve nlp-lab5-8-gpt2 (124M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 0.2 GB | 0.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 18, 2026.
nlp-lab5-8-gpt2 on every accelerator the SAVRN Index prices, at every precision
Model Card
The publisher has not written a card for this model.
Configuration
- Architecture
- GPT2LMHeadModel
- Vocabulary size
- 50,267
- Model type
- gpt2
Identity and Version
- Repository
- Tewin330/nlp-lab5-8-gpt2
- Publisher
- Fushi
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 124M parameters
- Languages
- Not stated by the source
- Revision
- c06da809d8ce0f070319ce0ef7be2b835543396f
- First published
- 2026-10-09
- Last updated
- 2026-10-09
Files and Weights
27 files, 3.5 GB in total. The weights are 12 files totalling 3.5 GB in bin, pt, pth, safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| checkpoint-40/model.safetensors | Weights | 497.8 MB | 80c3e58150fe |
| checkpoint-40/optimizer.pt | Weights | 995.7 MB | d5ac76625c63 |
| checkpoint-40/rng_state.pth | Weights | 14.5 KB | 4dc8c6913278 |
| checkpoint-40/scheduler.pt | Weights | 1.5 KB | 5525448f5b63 |
| checkpoint-40/training_args.bin | Weights | 5.2 KB | a7f8e6ec13f5 |
| checkpoint-60/model.safetensors | Weights | 497.8 MB | 1cb82904a217 |
| checkpoint-60/optimizer.pt | Weights | 995.7 MB | 3d3f1f81a816 |
| checkpoint-60/rng_state.pth | Weights | 14.5 KB | b90bfb5f2441 |
| checkpoint-60/scheduler.pt | Weights | 1.5 KB | 20844a8d67ae |
| checkpoint-60/training_args.bin | Weights | 5.2 KB | a7f8e6ec13f5 |
| model.safetensors | Weights | 497.8 MB | 1cb82904a217 |
| training_args.bin | Weights | 5.2 KB | a7f8e6ec13f5 |
| checkpoint-40/config.json | Configuration | 998 B | — |
| checkpoint-40/generation_config.json | Configuration | 140 B | — |
| checkpoint-40/trainer_state.json | Configuration | 7.4 KB | — |
| checkpoint-60/config.json | Configuration | 998 B | — |
| checkpoint-60/generation_config.json | Configuration | 140 B | — |
| checkpoint-60/trainer_state.json | Configuration | 10.7 KB | — |
| config.json | Configuration | 998 B | — |
| generation_config.json | Configuration | 140 B | — |
| .gitattributes | Repository | 1.5 KB | — |
| checkpoint-40/tokenizer.json | Tokenizer | 4.5 MB | — |
| checkpoint-40/tokenizer_config.json | Tokenizer | 347 B | — |
| checkpoint-60/tokenizer.json | Tokenizer | 4.5 MB | — |
| checkpoint-60/tokenizer_config.json | Tokenizer | 347 B | — |
| tokenizer.json | Tokenizer | 4.5 MB | — |
| tokenizer_config.json | Tokenizer | 347 B | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 3.5 GB
Released by Fushi through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 3.5 GB |
| 16-bit | 0.2 GB |
| 8-bit | 0.1 GB |
| 4-bit | 0.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About nlp-lab5-8-gpt2
How much GPU memory does nlp-lab5-8-gpt2 need?
About 0.3 GB at 16-bit and 0.1 GB at 4-bit: the weights (124M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run nlp-lab5-8-gpt2 on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.