stvg-r1-model-7b is an open-weight model from Javion, released under Creative Commons Attribution-NonCommercial-ShareAlike 4.0. It has 8.3B parameters and a 128,000-token context. At 16-bit it needs about 19.9 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 checkpoint for STVG-R1, the first reinforcement learning framework for spatial–temporal video grounding. We finetuned Qwen2.5-VL-7B-Instruct with constructed Visual Prompt Data.
Runs On
What it takes to serve stvg-r1-model-7b (8.3B 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 | 16.6 GB | 19.9 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 8.3 GB | 10.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 4.1 GB | 5.0 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.
stvg-r1-model-7b on every accelerator the SAVRN Index prices, at every precision
Model Card
This repository contains the checkpoint for STVG-R1, the first reinforcement learning framework for spatial–temporal video grounding. We finetuned Qwen2.5-VL-7B-Instruct with constructed Visual Prompt Data. If you use STVG-R1 or this checkpoint in your research, please cite
Excerpt from the card by Javion, licensed cc-by-nc-sa-4.0.
Configuration
- Architecture
- Qwen2_5_VLForConditionalGeneration
- Context length (tokens)
- 128,000
- Layers
- 28
- Hidden size
- 3,584
- Feed-forward size
- 18,944
- Attention heads
- 28
- Key/value heads
- 4
- Vocabulary size
- 152,064
- Sliding window (tokens)
- 32,768
- RoPE base
- 1e+06
- Stored precision
- bfloat16
- Model type
- qwen2_5_vl
Identity and Version
- Repository
- javion2025/stvg-r1-model-7b
- Publisher
- Javion
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 8.3B parameters
- Languages
- en
- Revision
- 8135b8463ae54bd4a48da3603f35bd3e3d502833
- First published
- 2026-09-23
- Last updated
- 2026-09-23
Files and Weights
19 files, 16.6 GB in total. The weights are 5 files totalling 16.6 GB in bin, safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00004.safetensors | Weights | 5.0 GB | 4bc52ba3c43b |
| model-00002-of-00004.safetensors | Weights | 5.0 GB | 764fdc53859f |
| model-00003-of-00004.safetensors | Weights | 4.9 GB | d93529f7cea3 |
| model-00004-of-00004.safetensors | Weights | 1.7 GB | ac060578a255 |
| training_args.bin | Weights | 8.8 KB | a278ea403081 |
| added_tokens.json | Configuration | 605 B | — |
| chat_template.json | Configuration | 1.0 KB | — |
| config.json | Configuration | 1.5 KB | — |
| generation_config.json | Configuration | 342 B | — |
| model.safetensors.index.json | Configuration | 57.6 KB | — |
| preprocessor_config.json | Configuration | 575 B | — |
| special_tokens_map.json | Configuration | 613 B | — |
| trainer_state.json | Configuration | 1.4 MB | — |
| README.md | Documentation | 2.0 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| merges.txt | Tokenizer | 1.7 MB | — |
| tokenizer.json | Tokenizer | 11.4 MB | 5eee858c5123 |
| tokenizer_config.json | Tokenizer | 7.3 KB | — |
| vocab.json | Tokenizer | 2.8 MB | — |
License and Download
- License
- cc-by-nc-sa-4.0
- Access
- Open weights, no gate
- Download size
- 16.6 GB
Released by Javion through its official repository on Hugging Face.
Built From
- Derived from Qwen/Qwen2.5-VL-7B-Instruct
- Described by arXiv:2602.11730
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 16.6 GB |
| 16-bit | 16.6 GB |
| 8-bit | 8.3 GB |
| 4-bit | 4.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About stvg-r1-model-7b
How much GPU memory does stvg-r1-model-7b need?
About 19.9 GB at 16-bit and 5 GB at 4-bit: the weights (8.3B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run stvg-r1-model-7b 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.
Can I use stvg-r1-model-7b commercially?
Not without separate permission. stvg-r1-model-7b is released under Creative Commons Attribution-NonCommercial-ShareAlike 4.0. CC BY-NC-SA 4.0 permits non-commercial sharing and adapting with credit, and requires adaptations to use the same license. Commercial use needs separate permission.
What is stvg-r1-model-7b's context length?
128,000 tokens, from the maximum position embeddings in its published configuration.