Open-weight model
visualrl-qwen3vl4b-ppo1-dapo-r2-fix
by Zeyuan Yang Miiche/visualrl-qwen3vl4b-ppo1-dapo-r2-fix
Parameters—
Context—
Weights128.9 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-dapo-r2-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
- 7e6067ff0c1b2f918409af48537e5e7e53ab761c
- First published
- 2026-09-16
- Last updated
- 2026-09-18
Files and Weights
95 files, 129.0 GB in total. The weights are 59 files totalling 128.9 GB in pt.
Weights59 files · 128.9 GB
Configuration18 files · 14.2 KB
Tokenizer12 files · 47.6 MB
Other5 files · 4.3 MB
Repository1 file · 2.4 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| global_step_80/actor/extra_state_world_size_8_rank_0.pt | Weights | 15.1 KB | 6ad3f4e7d3c2 |
| global_step_80/actor/extra_state_world_size_8_rank_1.pt | Weights | 15.1 KB | 7c3afbd00718 |
| global_step_80/actor/extra_state_world_size_8_rank_2.pt | Weights | 15.1 KB | 46ce6b976ee5 |
| global_step_80/actor/extra_state_world_size_8_rank_3.pt | Weights | 15.1 KB | 56cbee7c9260 |
| global_step_80/actor/extra_state_world_size_8_rank_4.pt | Weights | 15.1 KB | 6131a1e30bc5 |
| global_step_80/actor/extra_state_world_size_8_rank_5.pt | Weights | 15.1 KB | dc5d0f6a61cd |
| global_step_80/actor/extra_state_world_size_8_rank_6.pt | Weights | 15.1 KB | 4d306e4bd664 |
| global_step_80/actor/extra_state_world_size_8_rank_7.pt | Weights | 15.1 KB | 6d83fb1f09dc |
| global_step_80/actor/model_world_size_8_rank_0.pt | Weights | 2.4 GB | fa6fc07aab8f |
| global_step_80/actor/model_world_size_8_rank_1.pt | Weights | 2.4 GB | 9c899bb972f5 |
| global_step_80/actor/model_world_size_8_rank_2.pt | Weights | 2.4 GB | c3eda21855cb |
| global_step_80/actor/model_world_size_8_rank_3.pt | Weights | 2.4 GB | 704d0c884c0b |
| global_step_80/actor/model_world_size_8_rank_4.pt | Weights | 2.4 GB | c3a4d1c59bb7 |
| global_step_80/actor/model_world_size_8_rank_5.pt | Weights | 2.4 GB | 78093eedb5ab |
| global_step_80/actor/model_world_size_8_rank_6.pt | Weights | 2.4 GB | 83ba78ae677e |
| global_step_80/actor/model_world_size_8_rank_7.pt | Weights | 2.4 GB | be3682ebc6f5 |
| global_step_80/actor/optim_world_size_8_rank_0.pt | Weights | 4.4 GB | 14b660ba2c5c |
| global_step_80/actor/optim_world_size_8_rank_1.pt | Weights | 4.4 GB | 563dd9cccdfe |
| global_step_80/actor/optim_world_size_8_rank_2.pt | Weights | 4.4 GB | fba3e65a66a5 |
| global_step_80/actor/optim_world_size_8_rank_3.pt | Weights | 4.4 GB | d761515dfa50 |
| global_step_80/actor/optim_world_size_8_rank_4.pt | Weights | 4.4 GB | 8ed4c2e4828f |
| global_step_80/actor/optim_world_size_8_rank_5.pt | Weights | 4.4 GB | 0db06268aa96 |
| global_step_80/actor/optim_world_size_8_rank_6.pt | Weights | 4.4 GB | 44e7b67fa5e7 |
| global_step_80/actor/optim_world_size_8_rank_7.pt | Weights | 4.4 GB | 0f87e1c63979 |
| global_step_80/dataloader.pt | Weights | 2.0 KB | fd7580969d76 |
| global_step_90/actor/extra_state_world_size_8_rank_0.pt | Weights | 15.1 KB | 760f13a304a9 |
| global_step_90/actor/extra_state_world_size_8_rank_1.pt | Weights | 15.1 KB | 83acfe9c6fac |
| global_step_90/actor/extra_state_world_size_8_rank_2.pt | Weights | 15.1 KB | 181d82b502f4 |
| global_step_90/actor/extra_state_world_size_8_rank_3.pt | Weights | 15.1 KB | 1a9e29da286b |
| global_step_90/actor/extra_state_world_size_8_rank_4.pt | Weights | 15.1 KB | d0724d72ee50 |
| global_step_90/actor/extra_state_world_size_8_rank_5.pt | Weights | 15.1 KB | b7e69923a7e6 |
| global_step_90/actor/extra_state_world_size_8_rank_6.pt | Weights | 15.1 KB | 039cb36a9643 |
| global_step_90/actor/extra_state_world_size_8_rank_7.pt | Weights | 15.1 KB | 968346d925d6 |
| global_step_90/actor/model_world_size_8_rank_0.pt | Weights | 2.4 GB | 03c12d2c63a3 |
| global_step_90/actor/model_world_size_8_rank_1.pt | Weights | 2.4 GB | d06e00db4dca |
| global_step_90/actor/model_world_size_8_rank_2.pt | Weights | 2.4 GB | d66a8689cd0d |
| global_step_90/actor/model_world_size_8_rank_3.pt | Weights | 2.4 GB | ca0f2a8bf423 |
| global_step_90/actor/model_world_size_8_rank_4.pt | Weights | 2.4 GB | e44909399f46 |
| global_step_90/actor/model_world_size_8_rank_5.pt | Weights | 2.4 GB | 76ed24f499ce |
| global_step_90/actor/model_world_size_8_rank_6.pt | Weights | 2.4 GB | 9754dfec2528 |
| global_step_90/actor/model_world_size_8_rank_7.pt | Weights | 2.4 GB | 65755f687f29 |
| global_step_90/actor/optim_world_size_8_rank_0.pt | Weights | 4.4 GB | 1296a9555b2e |
| global_step_90/actor/optim_world_size_8_rank_1.pt | Weights | 4.4 GB | 53befc91a35a |
| global_step_90/actor/optim_world_size_8_rank_2.pt | Weights | 4.4 GB | 6f02e727e7f9 |
| global_step_90/actor/optim_world_size_8_rank_3.pt | Weights | 4.4 GB | 20addc67a2ab |
| global_step_90/actor/optim_world_size_8_rank_4.pt | Weights | 4.4 GB | 2549bab09679 |
| global_step_90/actor/optim_world_size_8_rank_5.pt | Weights | 4.4 GB | 61ea8ff5f67e |
| global_step_90/actor/optim_world_size_8_rank_6.pt | Weights | 4.4 GB | 317ba2d53615 |
| global_step_90/actor/optim_world_size_8_rank_7.pt | Weights | 4.4 GB | 754036c606ec |
| global_step_90/dataloader.pt | Weights | 2.0 KB | a8269a5fb1b8 |
| global_step_90_highlight/actor/model_world_size_8_rank_0.pt | Weights | 2.4 GB | 03c12d2c63a3 |
| global_step_90_highlight/actor/model_world_size_8_rank_1.pt | Weights | 2.4 GB | d06e00db4dca |
| global_step_90_highlight/actor/model_world_size_8_rank_2.pt | Weights | 2.4 GB | d66a8689cd0d |
| global_step_90_highlight/actor/model_world_size_8_rank_3.pt | Weights | 2.4 GB | ca0f2a8bf423 |
| global_step_90_highlight/actor/model_world_size_8_rank_4.pt | Weights | 2.4 GB | e44909399f46 |
| global_step_90_highlight/actor/model_world_size_8_rank_5.pt | Weights | 2.4 GB | 76ed24f499ce |
| global_step_90_highlight/actor/model_world_size_8_rank_6.pt | Weights | 2.4 GB | 9754dfec2528 |
| global_step_90_highlight/actor/model_world_size_8_rank_7.pt | Weights | 2.4 GB | 65755f687f29 |
| global_step_90_highlight/dataloader.pt | Weights | 2.0 KB | a8269a5fb1b8 |
| global_step_80/actor/huggingface/added_tokens.json | Configuration | 707 B | — |
| global_step_80/actor/huggingface/config.json | Configuration | 1.6 KB | — |
| global_step_80/actor/huggingface/generation_config.json | Configuration | 213 B | — |
| global_step_80/actor/huggingface/preprocessor_config.json | Configuration | 782 B | — |
| global_step_80/actor/huggingface/special_tokens_map.json | Configuration | 613 B | — |
| global_step_80/actor/huggingface/video_preprocessor_config.json | Configuration | 817 B | — |
| global_step_90/actor/huggingface/added_tokens.json | Configuration | 707 B | — |
| global_step_90/actor/huggingface/config.json | Configuration | 1.6 KB | — |
| global_step_90/actor/huggingface/generation_config.json | Configuration | 213 B | — |
| global_step_90/actor/huggingface/preprocessor_config.json | Configuration | 782 B | — |
| global_step_90/actor/huggingface/special_tokens_map.json | Configuration | 613 B | — |
| global_step_90/actor/huggingface/video_preprocessor_config.json | Configuration | 817 B | — |
| global_step_90_highlight/actor/huggingface/added_tokens.json | Configuration | 707 B | — |
| global_step_90_highlight/actor/huggingface/config.json | Configuration | 1.6 KB | — |
| global_step_90_highlight/actor/huggingface/generation_config.json | Configuration | 213 B | — |
| global_step_90_highlight/actor/huggingface/preprocessor_config.json | Configuration | 782 B | — |
| global_step_90_highlight/actor/huggingface/special_tokens_map.json | Configuration | 613 B | — |
| global_step_90_highlight/actor/huggingface/video_preprocessor_config.json | Configuration | 817 B | — |
| global_step_80/actor/huggingface/chat_template.jinja | Other | 5.3 KB | — |
| global_step_90/actor/huggingface/chat_template.jinja | Other | 5.3 KB | — |
| global_step_90_highlight/actor/huggingface/chat_template.jinja | Other | 5.3 KB | — |
| logs/eval-train-q3vl4b_ppo1_dapo_r2.log | Other | 4.1 MB | 17845bab7c38 |
| train_logs/train_metrics_q3vl4b_ppo1_dapo_r2.jsonl | Other | 159.6 KB | — |
| .gitattributes | Repository | 2.4 KB | — |
| global_step_80/actor/huggingface/merges.txt | Tokenizer | 1.7 MB | — |
| global_step_80/actor/huggingface/tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a |
| global_step_80/actor/huggingface/tokenizer_config.json | Tokenizer | 5.5 KB | — |
| global_step_80/actor/huggingface/vocab.json | Tokenizer | 2.8 MB | — |
| global_step_90/actor/huggingface/merges.txt | Tokenizer | 1.7 MB | — |
| global_step_90/actor/huggingface/tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a |
| global_step_90/actor/huggingface/tokenizer_config.json | Tokenizer | 5.5 KB | — |
| global_step_90/actor/huggingface/vocab.json | Tokenizer | 2.8 MB | — |
| global_step_90_highlight/actor/huggingface/merges.txt | Tokenizer | 1.7 MB | — |
| global_step_90_highlight/actor/huggingface/tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a |
| global_step_90_highlight/actor/huggingface/tokenizer_config.json | Tokenizer | 5.5 KB | — |
| global_step_90_highlight/actor/huggingface/vocab.json | Tokenizer | 2.8 MB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 128.9 GB
Download from Zeyuan Yang
Released by Zeyuan Yang through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 128.9 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.