Official implementation of UnifiedReward & [NeurIPS 2025] UnifiedReward-Think & UnifiedReward-Flex
active 2025-03-09 → 2026-06-18 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 467 days from 2025-03-09 to 2026-06-18. Pushes: 153 total, peak 13 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 76 total, peak 10 in a day. Comments: 102 total, peak 8 in a day. Stars: 266 total, peak 25 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| CodeGoat24 | 224 | 153 | 0 | 55 |
| kabalao | 9 | 0 | 0 | 4 |
| Henglin-Liu | 6 | 0 | 0 | 3 |
| MengHao666 | 6 | 0 | 0 | 3 |
| STAR-CV | 5 | 0 | 0 | 3 |
| stardusts-hj | 4 | 0 | 0 | 2 |
| duxiaodan | 4 | 0 | 0 | 2 |
| darkliang | 4 | 0 | 0 | 3 |
| zhang123434 | 4 | 0 | 0 | 1 |
| pmj110119 | 4 | 0 | 0 | 1 |
| ArlenCHEN | 3 | 0 | 0 | 2 |
| youyuanyi | 3 | 0 | 0 | 2 |
| wyn1015 | 3 | 0 | 0 | 1 |
| ZhuFengdaaa | 2 | 0 | 0 | 0 |
| g1aces | 2 | 0 | 0 | 0 |
| zgp123-wq | 2 | 0 | 0 | 1 |
| SuperKenVery | 2 | 0 | 0 | 0 |
| wwfnb | 2 | 0 | 0 | 0 |
| TimZhang001 | 2 | 0 | 0 | 1 |
| Sainzerjj | 2 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue#83CodeGoat242026-03-04 03:05UR-Flex seems to have a positional bias
- Issue comment#84youyuanyi2026-02-10 08:54请问下 unifiedreward-flex 需要的环境
- Issue comment#84youyuanyi2026-02-10 08:26请问下 unifiedreward-flex 需要的环境
- Issue comment#84CodeGoat242026-02-10 08:11请问下 unifiedreward-flex 需要的环境
- Issue comment#82CodeGoat242026-02-08 15:45nan values appearing in Style Scores when using UnifiedReward-2.0-qwen3vl-32b
- Issue#82Alisaleelee2026-02-08 15:35nan values appearing in Style Scores when using UnifiedReward-2.0-qwen3vl-32b
- Issue#81qianyue462-commits2026-01-28 13:58UnifiedReward-Think-qwen2.5vl-7b Model about pointwise inference setting
- Issue#78CodeGoat242025-12-28 07:46UnifiedReward-Think-qwen3vl相关模型支持pointwise打分吗
- Issue comment#76huan0851282025-12-24 04:53UnifiedReward-qwen3vl-8b测试效果问题
- Issue#77CodeGoat242025-12-11 01:20使用CodeGoat24/UnifiedReward-Think-qwen-7b是否必须resize到512*512
- Issue#77MengHao6662025-12-10 17:43使用CodeGoat24/UnifiedReward-Think-qwen-7b是否必须resize到512*512
- Issue comment#76huan0851282025-12-09 11:24UnifiedReward-qwen3vl-8b测试效果问题
- Issue#38CodeGoat242025-12-09 08:25Request for example data of grpo_train_data.json and rejection_sampling_train_data.json
- Issue#22CodeGoat242025-12-09 08:16the train config file of llama factory
- Issue#51CodeGoat242025-12-09 08:15Image and Video resizing
- Issue#40CodeGoat242025-12-09 08:03About data arrangement
- Issue#41CodeGoat242025-12-09 08:02Where is the GRPO training configuration?
- Issue#50CodeGoat242025-12-09 07:54image edit score
- Issue#26CodeGoat242025-12-09 07:52qwen_point_score_ACS_image_generation测试效果问题
- Issue#11CodeGoat242025-12-09 07:51Worse quality than sdxl-turbo
- Issue#47CodeGoat242025-12-09 07:421k训练数据在3B模型上的训练效果
- Issue#61CodeGoat242025-12-09 07:40Unified Reward如何在GenAI-Bench上测试？
- Issue comment#76CodeGoat242025-12-09 06:32测试发现qwen3 vl 8b效果不太好
- Issue comment#75MengHao6662025-12-07 07:43Does only "UnifiedReward-7b-v1.5" do point-score for single image?
- Issue#75MengHao6662025-12-07 03:19Does only "UnifiedReward-7b-v1.5" do point-score for single image?
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 266 stars here means stars gained during the window, not the repo's star count.