This repository contains the model-only checkpoint saved at 30,000 training steps. The files at the repository root are the original contents of checkpoint-30000. Full model and processor settings are in config.json, processorconfig.json, and experimentcfg/. The training run did not include an evaluation. Source checkpoint: outputs/moeobs2h10fourtasks20260926191812/pair67/usb/moe/checkpoint-30000.
Open-weight model · Robotics
GR00T-N1.7-ForceBenchmark-MoE-vasev6-MLP-30K-20260926
by Linqi Han HanLinqi/GR00T-N1.7-ForceBenchmark-MoE-vasev6-MLP-30K-20260926
GR00T-N1.7-ForceBenchmark-MoE-vasev6-MLP-30K-20260926 is an open-weight model for robotics from Linqi Han. It has 3.2B parameters. At 16-bit it needs about 7.6 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
This repository contains the model-only checkpoint saved at 30,000 training steps. The files at the repository root are the original contents of checkpoint-30000. Full model and processor settings are in config.json, processorconfig.json, and experimentcfg/.
Parameters3.2B
Context—
Weights7.0 GB
License—
AccessOpen weights
Monthly Downloads—
Runs On
What it takes to serve GR00T-N1.7-ForceBenchmark-MoE-vasev6-MLP-30K-20260926 (3.2B 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 | 6.3 GB | 7.6 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 3.2 GB | 3.8 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 1.6 GB | 1.9 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.
Model Card
This repository contains the model-only checkpoint saved at 30,000 training steps. The files at the repository root are the original contents of checkpoint-30000. Full model and processor settings are in config.json, processorconfig.json, and experimentcfg/. The training run did not include an evaluation. Source checkpoint: outputs/moeobs2h10fourtasks20260926191812/pair67/vasev6/moe/checkpoint-30000.
Excerpt from the card by Linqi Han.
Configuration
- Architecture
- Gr00tN1d7
- Hidden size
- 1,024
- Model type
- Gr00tN1d7
Identity and Version
- Repository
- HanLinqi/GR00T-N1.7-ForceBenchmark-MoE-vasev6-MLP-30K-20260926
- Publisher
- Linqi Han
- Task
- Robotics
- Modality
- Control
- Library
- Not stated by the source
- Parameters
- 3.2B parameters
- Languages
- moe
- Revision
- e2ab4f0b107b0dfe3e1b467e05ec04002f6ec356
- First published
- 2026-09-27
- Last updated
- 2026-09-27
Files and Weights
17 files, 7.0 GB in total. The weights are 3 files totalling 7.0 GB in bin, safetensors.
Weights3 files · 7.0 GB
Configuration12 files · 651.2 KB
Documentation1 file · 927 B
Repository1 file · 1.5 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00002.safetensors | Weights | 5.0 GB | f3ff92b5699c |
| model-00002-of-00002.safetensors | Weights | 2.0 GB | 066b8e2299a6 |
| training_args.bin | Weights | 8.0 KB | a46b3902384a |
| config.json | Configuration | 2.9 KB | — |
| embodiment_id.json | Configuration | 2.2 KB | — |
| experiment_cfg/conf.yaml | Configuration | 6.0 KB | — |
| experiment_cfg/config.yaml | Configuration | 5.9 KB | — |
| experiment_cfg/dataset_statistics.json | Configuration | 6.6 KB | — |
| experiment_cfg/final_model_config.json | Configuration | 2.2 KB | — |
| experiment_cfg/final_processor_config.json | Configuration | 12.5 KB | — |
| model.safetensors.index.json | Configuration | 182.9 KB | — |
| processor_config.json | Configuration | 3.1 KB | — |
| statistics.json | Configuration | 6.6 KB | — |
| trainer_state.json | Configuration | 420.1 KB | — |
| wandb_config.json | Configuration | 51 B | — |
| README.md | Documentation | 927 B | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 7.0 GB
Download from Linqi Han
Released by Linqi Han through its official repository on Hugging Face.
Built From
- Derived from nvidia/GR00T-N1.7-3B
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 7.0 GB |
| 16-bit | 6.3 GB |
| 8-bit | 3.2 GB |
| 4-bit | 1.6 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About GR00T-N1.7-ForceBenchmark-MoE-vasev6-MLP-30K-20260926
How much GPU memory does GR00T-N1.7-ForceBenchmark-MoE-vasev6-MLP-30K-20260926 need?
About 7.6 GB at 16-bit and 1.9 GB at 4-bit: the weights (3.2B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run GR00T-N1.7-ForceBenchmark-MoE-vasev6-MLP-30K-20260926 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.
Similar Models
This repository contains the model-only checkpoint saved at 30,000 training steps. The files at the repository root are the original contents of checkpoint-30000. Full model and processor settings are in config.json, processorconfig.json, and experimentcfg/. The training run did not include an evaluation. Source checkpoint: outputs/moeobs2h10fourtasks20260926191812/pair45/whiteboardv9/moe/checkpoint-30000.
This repository contains the model-only checkpoint saved at 30,000 training steps. The files at the repository root are the original contents of checkpoint-30000. Full model and processor settings are in config.json, processorconfig.json, and experimentcfg/. The training run did not include an evaluation. Source checkpoint: outputs/moeobs2h10fourtasks20260926191812/pair45/weightsorting/moe/checkpoint-30000.
This repository contains the model-only checkpoint saved at 30,000 training steps. The files at the repository root are the original contents of checkpoint-30000. Full model and processor settings are in config.json, processorconfig.json, and experimentcfg/. The training run did not include an evaluation. Source checkpoint: outputs/moeforcetransformertuningobs2h1020260926223332/usbk6s1e4f384xformer/usb/moe/checkpoint-30000.
This repository contains the model-only checkpoint saved at 30,000 training steps. The files at the repository root are the original contents of checkpoint-30000. Full model and processor settings are in config.json, processorconfig.json, and experimentcfg/. The training run did not include an evaluation. Source checkpoint: outputs/moeforcetransformertuningobs2h1020260926223332/whiteboardv9k8s1e6f384xformer/whiteboardv9/moe/checkpoint-30000.
This repository contains the model-only checkpoint saved at 30,000 training steps. The files at the repository root are the original contents of checkpoint-30000. Full model and processor settings are in config.json, processorconfig.json, and experimentcfg/. The training run did not include an evaluation. Source checkpoint: outputs/moeforcetransformertuningobs2h1020260926223332/vasev6k12s1e6f384xformer/vasev6/moe/checkpoint-30000.