Open-weight model
visualrl-qwen3vl4b-ppo1-4bx-c02-fix
by Zeyuan Yang Miiche/visualrl-qwen3vl4b-ppo1-4bx-c02-fix
visualrl-qwen3vl4b-ppo1-4bx-c02-fix is an open-weight model from Zeyuan Yang. Its published files total 55.1 GB.
Parameters—
Context—
Weights54.8 GB
License—
AccessOpen weights
Monthly Downloads—
Model Card
The publisher has not written a card for this model.
Identity and Version
- Repository
- Miiche/visualrl-qwen3vl4b-ppo1-4bx-c02-fix
- Publisher
- Zeyuan Yang
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- 4bf44f25436d033ba127f255a70af1c2d1216bf2
- First published
- 2026-09-20
- Last updated
- 2026-09-20
Files and Weights
27 files, 55.1 GB in total. The weights are 13 files totalling 54.8 GB in pt.
Weights13 files · 54.8 GB
Configuration6 files · 4.7 KB
Tokenizer4 files · 15.9 MB
Other3 files · 316.0 MB
Repository1 file · 1.7 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| global_step_10/actor/extra_state_world_size_4_rank_0.pt | Weights | 15.1 KB | 15b379b7a784 |
| global_step_10/actor/extra_state_world_size_4_rank_1.pt | Weights | 15.1 KB | 3c1eeff881e7 |
| global_step_10/actor/extra_state_world_size_4_rank_2.pt | Weights | 15.1 KB | 881dd6603493 |
| global_step_10/actor/extra_state_world_size_4_rank_3.pt | Weights | 15.1 KB | 7a1c647aacc7 |
| global_step_10/actor/model_world_size_4_rank_0.pt | Weights | 4.8 GB | 77a9debb20b5 |
| global_step_10/actor/model_world_size_4_rank_1.pt | Weights | 4.8 GB | 6c3a1ceb2129 |
| global_step_10/actor/model_world_size_4_rank_2.pt | Weights | 4.8 GB | 9aefe1fb2926 |
| global_step_10/actor/model_world_size_4_rank_3.pt | Weights | 4.8 GB | 09761433ea9c |
| global_step_10/actor/optim_world_size_4_rank_0.pt | Weights | 8.9 GB | 7a3533f8a062 |
| global_step_10/actor/optim_world_size_4_rank_1.pt | Weights | 8.9 GB | 3dce3b3c9e00 |
| global_step_10/actor/optim_world_size_4_rank_2.pt | Weights | 8.9 GB | ea8ce3272e78 |
| global_step_10/actor/optim_world_size_4_rank_3.pt | Weights | 8.9 GB | 135ddb06f907 |
| global_step_10/dataloader.pt | Weights | 2.0 KB | 5660d4aed01f |
| global_step_10/actor/huggingface/added_tokens.json | Configuration | 707 B | — |
| global_step_10/actor/huggingface/config.json | Configuration | 1.6 KB | — |
| global_step_10/actor/huggingface/generation_config.json | Configuration | 213 B | — |
| global_step_10/actor/huggingface/preprocessor_config.json | Configuration | 782 B | — |
| global_step_10/actor/huggingface/special_tokens_map.json | Configuration | 613 B | — |
| global_step_10/actor/huggingface/video_preprocessor_config.json | Configuration | 817 B | — |
| global_step_10/actor/huggingface/chat_template.jinja | Other | 5.3 KB | — |
| logs/eval-train-q3vl4b_ppo1_4bx_c02.log | Other | 316.0 MB | be04100c208b |
| train_logs/train_metrics_q3vl4b_ppo1_4bx_c02.jsonl | Other | 20.2 KB | — |
| .gitattributes | Repository | 1.7 KB | — |
| global_step_10/actor/huggingface/merges.txt | Tokenizer | 1.7 MB | — |
| global_step_10/actor/huggingface/tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a |
| global_step_10/actor/huggingface/tokenizer_config.json | Tokenizer | 5.5 KB | — |
| global_step_10/actor/huggingface/vocab.json | Tokenizer | 2.8 MB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 54.8 GB
Download from Zeyuan Yang
Released by Zeyuan Yang through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 54.8 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.