gr00t-n15-b601-can-bowl is an open-weight model from Rami Mouro. It has 2.7B parameters. At 16-bit it needs about 6.5 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
Parameters2.7B
Context—
Weights50.4 GB
License—
AccessOpen weights
Monthly Downloads—
Runs On
What it takes to serve gr00t-n15-b601-can-bowl (2.7B 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 | 5.4 GB | 6.5 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 2.7 GB | 3.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 1.4 GB | 1.6 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.
gr00t-n15-b601-can-bowl on every accelerator the SAVRN Index prices, at every precision
Model Card
The publisher has not written a card for this model.
Configuration
- Architecture
- GR00T_N1_5
- Hidden size
- 2,048
- Stored precision
- bfloat16
- Model type
- gr00t_n1_5
Identity and Version
- Repository
- ramalamadingdong/gr00t-n15-b601-can-bowl
- Publisher
- Rami Mouro
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 2.7B parameters
- Languages
- Not stated by the source
- Revision
- a634481a5f2ddf9fff339797e075fca3231a9761
- First published
- 2026-10-01
- Last updated
- 2026-10-02
Files and Weights
58 files, 59.7 GB in total. The weights are 21 files totalling 50.4 GB in bin, onnx, safetensors.
Weights21 files · 50.4 GB
Configuration26 files · 1.4 MB
Other10 files · 9.4 GB
Repository1 file · 2.0 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| checkpoint-10000/model-00001-of-00002.safetensors | Weights | 5.0 GB | aedf0adaffc7 |
| checkpoint-10000/model-00002-of-00002.safetensors | Weights | 2.6 GB | 86b020655fd7 |
| checkpoint-2000/model-00001-of-00002.safetensors | Weights | 5.0 GB | 480355b8b949 |
| checkpoint-2000/model-00002-of-00002.safetensors | Weights | 2.6 GB | d824f62b1370 |
| checkpoint-4000/model-00001-of-00002.safetensors | Weights | 5.0 GB | 53e91df34dc5 |
| checkpoint-4000/model-00002-of-00002.safetensors | Weights | 2.6 GB | 4cc2456ccd52 |
| checkpoint-6000/model-00001-of-00002.safetensors | Weights | 5.0 GB | 817ce5c12944 |
| checkpoint-6000/model-00002-of-00002.safetensors | Weights | 2.6 GB | 53e500346e71 |
| checkpoint-8000/model-00001-of-00002.safetensors | Weights | 5.0 GB | 86d9fb3ebc2e |
| checkpoint-8000/model-00002-of-00002.safetensors | Weights | 2.6 GB | 2766fe4dcc9e |
| export/checkpoint-10000/grootn15-qnn_context_binary-mixed-qualcomm_qcs9075/dit.bin | Weights | 870.9 MB | 65e4de880fab |
| export/checkpoint-10000/grootn15-qnn_context_binary-mixed-qualcomm_qcs9075/llm.bin | Weights | 632.8 MB | 54eb2fd25087 |
| export/checkpoint-10000/grootn15-qnn_context_binary-mixed-qualcomm_qcs9075/vit.bin | Weights | 430.4 MB | dc48e6a05d27 |
| export/checkpoint-10000/grootn15-qnn_context_binary-mixed-qualcomm_qcs9075/vlm_proj.bin | Weights | 256.5 MB | 157b5ee716b0 |
| model-00001-of-00002.safetensors | Weights | 5.0 GB | aedf0adaffc7 |
| model-00002-of-00002.safetensors | Weights | 2.6 GB | 86b020655fd7 |
| quant/checkpoint-10000/dit/model.onnx | Weights | 1.0 MB | a3a7c9f4d30a |
| quant/checkpoint-10000/llm/model.onnx | Weights | 345.7 KB | 2004d3e9eac6 |
| quant/checkpoint-10000/vit/model.onnx | Weights | 1.7 GB | 8d2561ba2bae |
| quant/checkpoint-10000/vlm_proj/model.onnx | Weights | 1.0 GB | 0a0cf896acc9 |
| training_args.bin | Weights | 5.3 KB | 20d2c4ff3369 |
| bundle/checkpoint-10000/bundle.json | Configuration | 5.7 KB | — |
| checkpoint-10000/config.json | Configuration | 1.7 KB | — |
| checkpoint-10000/experiment_cfg/metadata.json | Configuration | 7.0 KB | — |
| checkpoint-10000/model.safetensors.index.json | Configuration | 104.6 KB | — |
| checkpoint-10000/trainer_state.json | Configuration | 174.3 KB | — |
| checkpoint-2000/config.json | Configuration | 1.7 KB | — |
| checkpoint-2000/experiment_cfg/metadata.json | Configuration | 7.0 KB | — |
| checkpoint-2000/model.safetensors.index.json | Configuration | 104.6 KB | — |
| checkpoint-2000/trainer_state.json | Configuration | 34.9 KB | — |
| checkpoint-4000/config.json | Configuration | 1.7 KB | — |
| checkpoint-4000/experiment_cfg/metadata.json | Configuration | 7.0 KB | — |
| checkpoint-4000/model.safetensors.index.json | Configuration | 104.6 KB | — |
| checkpoint-4000/trainer_state.json | Configuration | 69.6 KB | — |
| checkpoint-6000/config.json | Configuration | 1.7 KB | — |
| checkpoint-6000/experiment_cfg/metadata.json | Configuration | 7.0 KB | — |
| checkpoint-6000/model.safetensors.index.json | Configuration | 104.6 KB | — |
| checkpoint-6000/trainer_state.json | Configuration | 104.4 KB | — |
| checkpoint-8000/config.json | Configuration | 1.7 KB | — |
| checkpoint-8000/experiment_cfg/metadata.json | Configuration | 7.0 KB | — |
| checkpoint-8000/model.safetensors.index.json | Configuration | 104.6 KB | — |
| checkpoint-8000/trainer_state.json | Configuration | 139.4 KB | — |
| config.json | Configuration | 1.7 KB | — |
| experiment_cfg/metadata.json | Configuration | 7.0 KB | — |
| export/checkpoint-10000/grootn15-qnn_context_binary-mixed-qualcomm_qcs9075/metadata.json | Configuration | 12.1 KB | — |
| model.safetensors.index.json | Configuration | 104.6 KB | — |
| trainer_state.json | Configuration | 174.6 KB | — |
| bundle/checkpoint-10000/text_embeds.f32 | Other | 5.2 MB | 5146fd970061 |
| export/checkpoint-10000/export.log | Other | 130.6 KB | — |
| quant/checkpoint-10000/dit/model.data | Other | 6.8 GB | 323caa6b12d2 |
| quant/checkpoint-10000/dit/model.encodings | Other | 48.8 MB | 720c5d3e004e |
| quant/checkpoint-10000/llm/model.data | Other | 2.5 GB | 09797f8762e9 |
| quant/checkpoint-10000/llm/model.encodings | Other | 15.8 MB | a8a355f50383 |
| quant/checkpoint-10000/quantize.log | Other | 171.7 KB | — |
| quant/checkpoint-10000/vit/model.encodings | Other | 17.3 MB | 12a1422c03e4 |
| quant/checkpoint-10000/vlm_proj/model.encodings | Other | 6.1 MB | — |
| runs/Oct01_18-11-16_j-ramalamadingdong-6abea165fbc85ba682363e66-t9dt7fnr-8b696xw52h/events.out.tfevents.1790878279.j-ramalamadingdong-6abea165fbc85ba682363e66-t9dt7fnr-8b696xw52h.3426.0 | Other | 217.4 KB | 51db11988803 |
| .gitattributes | Repository | 2.0 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 50.4 GB
Download from Rami Mouro
Released by Rami Mouro through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 50.4 GB |
| 16-bit | 5.4 GB |
| 8-bit | 2.7 GB |
| 4-bit | 1.4 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About gr00t-n15-b601-can-bowl
How much GPU memory does gr00t-n15-b601-can-bowl need?
About 6.5 GB at 16-bit and 1.6 GB at 4-bit: the weights (2.7B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run gr00t-n15-b601-can-bowl 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.