Open-weight model
visualrl-qwen3vl2b-ppo1-ral-c100-clean
by Zeyuan Yang Miiche/visualrl-qwen3vl2b-ppo1-ral-c100-clean
Parameters—
Context—
Weights14.6 GB
License—
AccessOpen weights
Monthly Downloads—
Model Card
The publisher has not written a card for this model.
Identity and Version
- Repository
- Miiche/visualrl-qwen3vl2b-ppo1-ral-c100-clean
- 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
- c7420fb41e71c887c4c8f87b5b3fe44871d3b50e
- First published
- 2026-09-18
- Last updated
- 2026-09-18
Files and Weights
155 files, 33.1 GB in total. The weights are 3 files totalling 14.6 GB in safetensors.
Weights3 files · 14.6 GB
Configuration94 files · 251.7 MB
Tokenizer12 files · 47.6 MB
Other45 files · 18.2 GB
Repository1 file · 2.8 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| global_step_180_merged/model.safetensors | Weights | 4.9 GB | ecb2dcd505d5 |
| global_step_270_merged/model.safetensors | Weights | 4.9 GB | 8eb369df31ab |
| global_step_90_merged/model.safetensors | Weights | 4.9 GB | a01e28076f8a |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/blink_output.json | Configuration | 9.2 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/blink_scores.json | Configuration | 3.2 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/chartqa_output.json | Configuration | 2.5 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/chartqa_scores.json | Configuration | 191 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/hallusionbench_output.json | Configuration | 2.4 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/hallusionbench_scores.json | Configuration | 493 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/logicvista_output.json | Configuration | 3.9 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/logicvista_scores.json | Configuration | 503 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mathverse_output.json | Configuration | 21.4 MB | af29790ea6e3 |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mathverse_scores.json | Configuration | 2.3 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mathvision_output.json | Configuration | 4.1 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mathvision_scores.json | Configuration | 8.8 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mathvista_output.json | Configuration | 5.4 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mathvista_scores.json | Configuration | 7.5 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmbench_output.json | Configuration | 12.7 MB | 74ca074f293c |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmbench_scores.json | Configuration | 6.4 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mme_output.json | Configuration | 2.9 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mme_scores.json | Configuration | 2.0 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmmu_output.json | Configuration | 6.2 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmmu_pro_output.json | Configuration | 196.3 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmmu_pro_scores.json | Configuration | 108 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmmu_scores.json | Configuration | 2.6 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmstar_output.json | Configuration | 6.4 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmstar_scores.json | Configuration | 3.1 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmvet_output.json | Configuration | 234.7 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmvet_scores.json | Configuration | 491 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/perceptionbench_output.json | Configuration | 23.9 MB | f7608d41c9c9 |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/perceptionbench_scores.json | Configuration | 1.0 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/realworldqa_output.json | Configuration | 1.2 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/realworldqa_scores.json | Configuration | 95 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/scienceqa_output.json | Configuration | 5.4 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/scienceqa_scores.json | Configuration | 359 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/visnumbench_output.json | Configuration | 11.8 MB | 87a986b32753 |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/visnumbench_scores.json | Configuration | 952 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/chartqa_output.json | Configuration | 2.4 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/chartqa_scores.json | Configuration | 191 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/mmvet_output.json | Configuration | 260.1 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/mmvet_scores.json | Configuration | 527 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/realworldqa_output.json | Configuration | 1.2 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/realworldqa_scores.json | Configuration | 95 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/visnumbench_output.json | Configuration | 9.6 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/visnumbench_scores.json | Configuration | 949 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/blink_output.json | Configuration | 8.4 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/blink_scores.json | Configuration | 3.2 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/chartqa_output.json | Configuration | 2.2 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/chartqa_scores.json | Configuration | 193 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/hallusionbench_output.json | Configuration | 2.3 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/hallusionbench_scores.json | Configuration | 492 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/logicvista_output.json | Configuration | 4.2 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/logicvista_scores.json | Configuration | 501 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mathverse_output.json | Configuration | 23.1 MB | 4ffa7a9e9e58 |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mathverse_scores.json | Configuration | 2.3 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mathvision_output.json | Configuration | 4.0 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mathvision_scores.json | Configuration | 8.8 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mathvista_output.json | Configuration | 5.0 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mathvista_scores.json | Configuration | 7.4 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmbench_output.json | Configuration | 12.7 MB | 3bf3fa601904 |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmbench_scores.json | Configuration | 6.4 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mme_output.json | Configuration | 2.7 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mme_scores.json | Configuration | 1.9 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmmu_output.json | Configuration | 6.2 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmmu_pro_output.json | Configuration | 196.3 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmmu_pro_scores.json | Configuration | 108 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmmu_scores.json | Configuration | 2.7 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmstar_output.json | Configuration | 6.3 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmstar_scores.json | Configuration | 3.1 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmvet_output.json | Configuration | 266.8 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmvet_scores.json | Configuration | 517 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/perceptionbench_output.json | Configuration | 22.2 MB | 1ef90029d3e0 |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/perceptionbench_scores.json | Configuration | 1.1 KB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/realworldqa_output.json | Configuration | 1.5 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/realworldqa_scores.json | Configuration | 95 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/scienceqa_output.json | Configuration | 5.5 MB | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/scienceqa_scores.json | Configuration | 359 B | — |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/visnumbench_output.json | Configuration | 11.4 MB | 9fe9f00f57ff |
| eval_results/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/visnumbench_scores.json | Configuration | 950 B | — |
| global_step_180_merged/added_tokens.json | Configuration | 707 B | — |
| global_step_180_merged/config.json | Configuration | 1.6 KB | — |
| global_step_180_merged/generation_config.json | Configuration | 147 B | — |
| global_step_180_merged/preprocessor_config.json | Configuration | 782 B | — |
| global_step_180_merged/special_tokens_map.json | Configuration | 613 B | — |
| global_step_180_merged/video_preprocessor_config.json | Configuration | 817 B | — |
| global_step_270_merged/added_tokens.json | Configuration | 707 B | — |
| global_step_270_merged/config.json | Configuration | 1.6 KB | — |
| global_step_270_merged/generation_config.json | Configuration | 147 B | — |
| global_step_270_merged/preprocessor_config.json | Configuration | 782 B | — |
| global_step_270_merged/special_tokens_map.json | Configuration | 613 B | — |
| global_step_270_merged/video_preprocessor_config.json | Configuration | 817 B | — |
| global_step_90_merged/added_tokens.json | Configuration | 707 B | — |
| global_step_90_merged/config.json | Configuration | 1.6 KB | — |
| global_step_90_merged/generation_config.json | Configuration | 147 B | — |
| global_step_90_merged/preprocessor_config.json | Configuration | 782 B | — |
| global_step_90_merged/special_tokens_map.json | Configuration | 613 B | — |
| global_step_90_merged/video_preprocessor_config.json | Configuration | 817 B | — |
| global_step_180_merged/chat_template.jinja | Other | 5.3 KB | — |
| global_step_270_merged/chat_template.jinja | Other | 5.3 KB | — |
| global_step_90_merged/chat_template.jinja | Other | 5.3 KB | — |
| logs/eval-train-q3vl2b_ppo1_ral_c100.log | Other | 18.2 GB | 0c80c771e8d4 |
| logs/eval-visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm.log | Other | 1.6 MB | — |
| logs/eval-visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm.log | Other | 1.6 MB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/blink.log | Other | 147.3 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/chartqa.log | Other | 94.8 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/hallusionbench.log | Other | 50.4 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/logicvista.log | Other | 34.0 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mathverse.log | Other | 198.7 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mathvision.log | Other | 26.8 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mathvista.log | Other | 60.1 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmbench.log | Other | 155.1 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mme.log | Other | 103.6 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmmu.log | Other | 84.2 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmmu_pro.log | Other | 116.5 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmstar.log | Other | 75.6 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/mmvet.log | Other | 44.1 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/perceptionbench.log | Other | 186.5 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/realworldqa.log | Other | 56.8 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/scienceqa.log | Other | 83.6 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s180-vllm/visnumbench.log | Other | 85.9 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/chartqa.log | Other | 95.7 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/mmvet.log | Other | 44.7 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/realworldqa.log | Other | 56.4 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s270-vllm/visnumbench.log | Other | 92.4 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/blink.log | Other | 147.2 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/chartqa.log | Other | 96.2 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/hallusionbench.log | Other | 49.7 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/logicvista.log | Other | 33.7 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mathverse.log | Other | 198.8 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mathvision.log | Other | 27.0 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mathvista.log | Other | 59.9 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmbench.log | Other | 162.7 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mme.log | Other | 101.5 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmmu.log | Other | 84.3 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmmu_pro.log | Other | 122.7 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmstar.log | Other | 74.3 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/mmvet.log | Other | 45.1 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/perceptionbench.log | Other | 186.6 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/realworldqa.log | Other | 61.1 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/scienceqa.log | Other | 84.9 KB | — |
| logs/visualrl-qwen3vl2b-ppo1-ral-c100-clean-s90-vllm/visnumbench.log | Other | 88.9 KB | — |
| train_logs/train_metrics_q3vl2b_ppo1_ral_c100.jsonl | Other | 542.3 KB | — |
| .gitattributes | Repository | 2.8 KB | — |
| global_step_180_merged/merges.txt | Tokenizer | 1.7 MB | — |
| global_step_180_merged/tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a |
| global_step_180_merged/tokenizer_config.json | Tokenizer | 5.5 KB | — |
| global_step_180_merged/vocab.json | Tokenizer | 2.8 MB | — |
| global_step_270_merged/merges.txt | Tokenizer | 1.7 MB | — |
| global_step_270_merged/tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a |
| global_step_270_merged/tokenizer_config.json | Tokenizer | 5.5 KB | — |
| global_step_270_merged/vocab.json | Tokenizer | 2.8 MB | — |
| global_step_90_merged/merges.txt | Tokenizer | 1.7 MB | — |
| global_step_90_merged/tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a |
| global_step_90_merged/tokenizer_config.json | Tokenizer | 5.5 KB | — |
| global_step_90_merged/vocab.json | Tokenizer | 2.8 MB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 14.6 GB
Download from Zeyuan Yang
Released by Zeyuan Yang through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 14.6 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.