Open-weight model
nepali-llama-590m-250m-mixed-cpt
by ZENITH Anurag9817/nepali-llama-590m-250m-mixed-cpt
nepali-llama-590m-250m-mixed-cpt is an open-weight model from ZENITH. It has 590M parameters and a 2,048-token context. At 16-bit it needs about 1.4 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 nepali-llama-590m-250m-mixed-cpt (590M 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 | 1.2 GB | 1.4 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.6 GB | 0.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.3 GB | 0.4 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.
nepali-llama-590m-250m-mixed-cpt on every accelerator the SAVRN Index prices, at every precision
Model Card
The publisher has not written a card for this model.
Configuration
- Architecture
- LlamaForCausalLM
- Context length (tokens)
- 2,048
- Layers
- 18
- Hidden size
- 1,536
- Feed-forward size
- 4,096
- Attention heads
- 12
- Key/value heads
- 6
- Head dimension
- 128
- Vocabulary size
- 40,000
- Model type
- llama
Identity and Version
- Repository
- Anurag9817/nepali-llama-590m-250m-mixed-cpt
- Publisher
- ZENITH
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 590M parameters
- Languages
- Not stated by the source
- Revision
- f30b5daa69bd222cb312844796881b220776f429
- First published
- 2026-10-05
- Last updated
- 2026-10-05
Files and Weights
7 files, 2.4 GB in total. The weights are 2 files totalling 2.4 GB in bin, safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 2.4 GB | c2c3dfda256c |
| training_args.bin | Weights | 5.3 KB | 3fe215bcbc3c |
| config.json | Configuration | 836 B | — |
| generation_config.json | Configuration | 102 B | — |
| training_config.json | Configuration | 366 B | — |
| nepali_llama_40k.model | Other | 938.5 KB | 79df19916cd8 |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 2.4 GB
Released by ZENITH through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 2.4 GB |
| 16-bit | 1.2 GB |
| 8-bit | 0.6 GB |
| 4-bit | 0.3 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About nepali-llama-590m-250m-mixed-cpt
How much GPU memory does nepali-llama-590m-250m-mixed-cpt need?
About 1.4 GB at 16-bit and 0.4 GB at 4-bit: the weights (590M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run nepali-llama-590m-250m-mixed-cpt 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.
What is nepali-llama-590m-250m-mixed-cpt's context length?
2,048 tokens, from the maximum position embeddings in its published configuration.