Open-weight model
dspark_grug-67b-a2b-datakit-sft_datakit-sft-mix_itw-20261016
by Atula Tejaswi atutej/dspark_grug-67b-a2b-datakit-sft_datakit-sft-mix_itw-20261016
dspark_grug-67b-a2b-datakit-sft_datakit-sft-mix_itw-20261016 is an open-weight model from Atula Tejaswi. It has 280M parameters. At 16-bit it needs about 0.7 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
A DSpark speculative-decoding drafter (block size 8, 3 layers, reduced draft vocabulary 32000) for the target model open-athena/Grug-67B-A2B-Datakit-SFT-262K-2026.09.21.
Runs On
What it takes to serve dspark_grug-67b-a2b-datakit-sft_datakit-sft-mix_itw-20261016 (280M 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 | 0.6 GB | 0.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.3 GB | 0.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.1 GB | 0.2 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 20, 2026.
Model Card
A DSpark speculative-decoding drafter (block size 8, 3 layers, reduced draft vocabulary 32000) for the target model open-athena/Grug-67B-A2B-Datakit-SFT-262K-2026.09.21. Trained with the speculators library (fork atutej/speculators, branch grug-dspark-gate); run name dsparkgrug-67b-a2b-datakit-sftdatakit-sft-mixitw. itw drafter (in-the-wild text): trained on the gold replies of the Datakit SFT mix, the data the target was fine-tuned on, as written and with no regeneration. This is a mid-run checkpoint at step 80,813 (21% of one epoch of 382,747 steps, about 9.7B tokens, 12 h on 2 GB200 nodes); training was stopped by a time limit before the learning rate had decayed. The trainer's…
Excerpt from the card by Atula Tejaswi.
Configuration
- Architecture
- DSparkDraftModel
Identity and Version
- Repository
- atutej/dspark_grug-67b-a2b-datakit-sft_datakit-sft-mix_itw-20261016
- Publisher
- Atula Tejaswi
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 280M parameters
- Languages
- Not stated by the source
- Revision
- 764de905927ce5295554d2e30e1c52620794906b
- First published
- 2026-10-17
- Last updated
- 2026-10-17
Files and Weights
7 files, 561.0 MB in total. The weights are 1 file totalling 560.9 MB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 560.9 MB | f528484f7108 |
| config.json | Configuration | 2.0 KB | — |
| config.py | Configuration | 2.3 KB | — |
| training_data_manifest.json | Configuration | 40.3 KB | — |
| training_state.json | Configuration | 55 B | — |
| README.md | Documentation | 12.1 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 560.9 MB
Released by Atula Tejaswi through its official repository on Hugging Face.
Built From
- Derived from open-athena/Grug-67B-A2B-Datakit-SFT-262K-2026.09.21
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 560.9 MB |
| 16-bit | 0.6 GB |
| 8-bit | 0.3 GB |
| 4-bit | 0.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About dspark_grug-67b-a2b-datakit-sft_datakit-sft-mix_itw-20261016
How much GPU memory does dspark_grug-67b-a2b-datakit-sft_datakit-sft-mix_itw-20261016 need?
About 0.7 GB at 16-bit and 0.2 GB at 4-bit: the weights (280M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run dspark_grug-67b-a2b-datakit-sft_datakit-sft-mix_itw-20261016 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.