Skip to page content
SAVRN
Search Plan a project

Open-weight model

visualrl-qwen3vl4b-ppo1-4cx-c100-clean

by Zeyuan Yang Miiche/visualrl-qwen3vl4b-ppo1-4cx-c100-clean

visualrl-qwen3vl4b-ppo1-4cx-c100-clean is an open-weight model from Zeyuan Yang. Its published files total 42.1 GB.

Parameters—
Context—
Weights29.0 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-ppo1-4cx-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
22ee8d149475e3fb8610428092b18463e3bb82e8
First published
2026-09-24
Last updated
2026-09-25

Files and Weights

202 files, 42.1 GB in total. The weights are 7 files totalling 29.0 GB in safetensors.

Weights7 files · 29.0 GB
Configuration123 files · 188.3 MB
Tokenizer12 files · 47.6 MB
Other59 files · 12.9 GB
Repository1 file · 2.3 KB
Every file
FileTypeSizeSHA-256
global_step_180_merged/model-00001-of-00002.safetensorsWeights5.0 GB b5235a86bc23
global_step_180_merged/model-00002-of-00002.safetensorsWeights4.7 GB ca27baadf3db
global_step_270_merged/model-00001-of-00003.safetensorsWeights4.3 GB 5700fc373079
global_step_270_merged/model-00002-of-00003.safetensorsWeights5.0 GB 545c57f604a3
global_step_270_merged/model-00003-of-00003.safetensorsWeights417.9 MB dd1ed0f99433
global_step_90_merged/model-00001-of-00002.safetensorsWeights5.0 GB b6b0a28dccb2
global_step_90_merged/model-00002-of-00002.safetensorsWeights4.7 GB bf7bd98b2f2f
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/blink_output.jsonConfiguration6.3 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/blink_scores.jsonConfiguration3.2 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/chartqa_output.jsonConfiguration3.2 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/chartqa_scores.jsonConfiguration193 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/hallusionbench_output.jsonConfiguration1.8 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/hallusionbench_scores.jsonConfiguration492 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/logicvista_output.jsonConfiguration2.6 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/logicvista_scores.jsonConfiguration501 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mathverse_output.jsonConfiguration14.4 MB fbcb6834b29c
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mathverse_scores.jsonConfiguration2.3 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mathvision_output.jsonConfiguration2.7 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mathvision_scores.jsonConfiguration8.9 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mathvista_output.jsonConfiguration3.4 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mathvista_scores.jsonConfiguration7.5 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmbench_output.jsonConfiguration8.7 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmbench_scores.jsonConfiguration6.3 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mme_output.jsonConfiguration3.6 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mme_scores.jsonConfiguration2.0 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmmu_output.jsonConfiguration5.0 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmmu_pro_output.jsonConfiguration196.1 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmmu_pro_scores.jsonConfiguration108 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmmu_scores.jsonConfiguration2.6 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmstar_output.jsonConfiguration4.3 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmstar_scores.jsonConfiguration3.1 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmvet_output.jsonConfiguration198.1 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmvet_scores.jsonConfiguration502 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/perceptionbench_output.jsonConfiguration16.4 MB 1456976c88b1
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/perceptionbench_scores.jsonConfiguration1.0 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/realworldqa_output.jsonConfiguration1.4 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/realworldqa_scores.jsonConfiguration95 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/scienceqa_output.jsonConfiguration3.0 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/scienceqa_scores.jsonConfiguration359 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/visnumbench_output.jsonConfiguration5.6 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/visnumbench_scores.jsonConfiguration952 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/blink_output.jsonConfiguration834.2 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/blink_scores.jsonConfiguration3.2 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/chartqa_output.jsonConfiguration1.3 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/chartqa_scores.jsonConfiguration193 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/hallusionbench_output.jsonConfiguration469.4 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/hallusionbench_scores.jsonConfiguration491 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/logicvista_output.jsonConfiguration1.7 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/logicvista_scores.jsonConfiguration500 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mathverse_output.jsonConfiguration3.5 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mathverse_scores.jsonConfiguration2.3 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mathvision_output.jsonConfiguration1.2 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mathvision_scores.jsonConfiguration8.9 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mathvista_output.jsonConfiguration1.7 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mathvista_scores.jsonConfiguration7.5 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmbench_output.jsonConfiguration3.0 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmbench_scores.jsonConfiguration6.4 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mme_output.jsonConfiguration679.3 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mme_scores.jsonConfiguration2.0 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmmu_output.jsonConfiguration944.6 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmmu_pro_output.jsonConfiguration196.2 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmmu_pro_scores.jsonConfiguration110 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmmu_scores.jsonConfiguration2.6 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmstar_output.jsonConfiguration1.2 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmstar_scores.jsonConfiguration3.1 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmvet_output.jsonConfiguration210.1 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmvet_scores.jsonConfiguration492 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/perceptionbench_output.jsonConfiguration1.1 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/perceptionbench_scores.jsonConfiguration1.1 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/realworldqa_output.jsonConfiguration222.4 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/realworldqa_scores.jsonConfiguration95 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/scienceqa_output.jsonConfiguration1.3 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/scienceqa_scores.jsonConfiguration359 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/visnumbench_output.jsonConfiguration1.1 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/visnumbench_scores.jsonConfiguration951 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/blink_output.jsonConfiguration6.1 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/blink_scores.jsonConfiguration3.2 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/chartqa_output.jsonConfiguration3.4 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/chartqa_scores.jsonConfiguration193 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/hallusionbench_output.jsonConfiguration2.0 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/hallusionbench_scores.jsonConfiguration492 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/logicvista_output.jsonConfiguration2.6 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/logicvista_scores.jsonConfiguration501 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mathverse_output.jsonConfiguration15.7 MB 769b079940c8
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mathverse_scores.jsonConfiguration2.3 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mathvision_output.jsonConfiguration2.7 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mathvision_scores.jsonConfiguration9.0 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mathvista_output.jsonConfiguration3.8 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mathvista_scores.jsonConfiguration7.5 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmbench_output.jsonConfiguration8.9 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmbench_scores.jsonConfiguration6.4 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mme_output.jsonConfiguration4.2 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mme_scores.jsonConfiguration2.0 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmmu_output.jsonConfiguration4.8 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmmu_pro_output.jsonConfiguration196.1 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmmu_pro_scores.jsonConfiguration110 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmmu_scores.jsonConfiguration2.7 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmstar_output.jsonConfiguration4.2 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmstar_scores.jsonConfiguration3.1 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmvet_output.jsonConfiguration215.2 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmvet_scores.jsonConfiguration501 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/perceptionbench_output.jsonConfiguration16.1 MB 71f861e0c1be
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/perceptionbench_scores.jsonConfiguration1.0 KB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/realworldqa_output.jsonConfiguration1.1 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/realworldqa_scores.jsonConfiguration94 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/scienceqa_output.jsonConfiguration3.4 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/scienceqa_scores.jsonConfiguration359 B —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/visnumbench_output.jsonConfiguration5.4 MB —
eval_results/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/visnumbench_scores.jsonConfiguration953 B —
global_step_180_merged/added_tokens.jsonConfiguration707 B —
global_step_180_merged/config.jsonConfiguration1.6 KB —
global_step_180_merged/generation_config.jsonConfiguration147 B —
global_step_180_merged/model.safetensors.index.jsonConfiguration64.8 KB —
global_step_180_merged/preprocessor_config.jsonConfiguration782 B —
global_step_180_merged/special_tokens_map.jsonConfiguration613 B —
global_step_180_merged/video_preprocessor_config.jsonConfiguration817 B —
global_step_270_merged/added_tokens.jsonConfiguration707 B —
global_step_270_merged/config.jsonConfiguration1.6 KB —
global_step_270_merged/generation_config.jsonConfiguration147 B —
global_step_270_merged/model.safetensors.index.jsonConfiguration64.8 KB —
global_step_270_merged/preprocessor_config.jsonConfiguration782 B —
global_step_270_merged/special_tokens_map.jsonConfiguration613 B —
global_step_270_merged/video_preprocessor_config.jsonConfiguration817 B —
global_step_90_merged/added_tokens.jsonConfiguration707 B —
global_step_90_merged/config.jsonConfiguration1.6 KB —
global_step_90_merged/generation_config.jsonConfiguration147 B —
global_step_90_merged/model.safetensors.index.jsonConfiguration64.8 KB —
global_step_90_merged/preprocessor_config.jsonConfiguration782 B —
global_step_90_merged/special_tokens_map.jsonConfiguration613 B —
global_step_90_merged/video_preprocessor_config.jsonConfiguration817 B —
global_step_180_merged/chat_template.jinjaOther5.3 KB —
global_step_270_merged/chat_template.jinjaOther5.3 KB —
global_step_90_merged/chat_template.jinjaOther5.3 KB —
logs/eval-train-q3vl4b_ppo1_4cx_c100.logOther12.9 GB 1848399b4233
logs/eval-visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm.logOther2.3 MB —
logs/eval-visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm.logOther1.3 MB —
logs/eval-visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm.logOther2.3 MB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/blink.logOther165.4 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/chartqa.logOther105.3 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/hallusionbench.logOther59.7 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/logicvista.logOther54.3 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mathverse.logOther308.9 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mathvision.logOther44.1 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mathvista.logOther79.4 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmbench.logOther186.6 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mme.logOther120.0 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmmu.logOther107.2 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmmu_pro.logOther166.6 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmstar.logOther94.7 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/mmvet.logOther174.6 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/perceptionbench.logOther258.9 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/realworldqa.logOther59.8 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/scienceqa.logOther98.3 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s180-vllm/visnumbench.logOther161.7 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/blink.logOther117.5 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/chartqa.logOther86.4 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/hallusionbench.logOther36.5 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/logicvista.logOther33.5 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mathverse.logOther138.8 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mathvision.logOther25.9 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mathvista.logOther50.0 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmbench.logOther90.9 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mme.logOther92.8 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmmu.logOther68.0 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmmu_pro.logOther66.9 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmstar.logOther50.4 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/mmvet.logOther166.7 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/perceptionbench.logOther111.2 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/realworldqa.logOther42.6 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/scienceqa.logOther74.9 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s270-vllm/visnumbench.logOther70.0 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/blink.logOther169.5 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/chartqa.logOther109.3 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/hallusionbench.logOther63.4 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/logicvista.logOther52.2 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mathverse.logOther311.6 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mathvision.logOther45.3 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mathvista.logOther79.8 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmbench.logOther196.3 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mme.logOther125.9 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmmu.logOther106.8 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmmu_pro.logOther169.8 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmstar.logOther96.2 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/mmvet.logOther165.6 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/perceptionbench.logOther284.9 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/realworldqa.logOther59.0 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/scienceqa.logOther100.9 KB —
logs/visualrl-qwen3vl4b-ppo1-4cx-c100-clean-s90-vllm/visnumbench.logOther172.8 KB —
train_logs/train_metrics_q3vl4b_ppo1_4cx_c100.jsonlOther543.7 KB —
.gitattributesRepository2.3 KB —
global_step_180_merged/merges.txtTokenizer1.7 MB —
global_step_180_merged/tokenizer.jsonTokenizer11.4 MB aeb13307a71a
global_step_180_merged/tokenizer_config.jsonTokenizer5.5 KB —
global_step_180_merged/vocab.jsonTokenizer2.8 MB —
global_step_270_merged/merges.txtTokenizer1.7 MB —
global_step_270_merged/tokenizer.jsonTokenizer11.4 MB aeb13307a71a
global_step_270_merged/tokenizer_config.jsonTokenizer5.5 KB —
global_step_270_merged/vocab.jsonTokenizer2.8 MB —
global_step_90_merged/merges.txtTokenizer1.7 MB —
global_step_90_merged/tokenizer.jsonTokenizer11.4 MB aeb13307a71a
global_step_90_merged/tokenizer_config.jsonTokenizer5.5 KB —
global_step_90_merged/vocab.jsonTokenizer2.8 MB —

License and Download

License
Not stated by the source
Access
Open weights, no gate
Download size
29.0 GB
Download from Zeyuan Yang

Released by Zeyuan Yang through its official repository on Hugging Face.

Memory Requirements

PrecisionWeights in memory
As published29.0 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.