[ICML 2026 Spotlight] WeDLM: The fastest diffusion language model with standard causal attention and native KV cache compatibility, delivering real speedups over vLLM-optimized baselines.
active 2025-12-29 → 2026-06-02 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 156 days from 2025-12-29 to 2026-06-02. Pushes: 4 total, peak 1 in a day. Pull requests: 6 total, peak 3 in a day. Issues: 17 total, peak 3 in a day. Comments: 25 total, peak 6 in a day. Stars: 260 total, peak 60 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 |
|---|---|---|---|---|
| exlaw | 23 | 4 | 0 | 12 |
| AIxyz | 5 | 0 | 0 | 2 |
| MatthewCYM | 3 | 0 | 0 | 2 |
| aakash-chaddha | 3 | 0 | 2 | 1 |
| yjyddq | 2 | 0 | 0 | 0 |
| neoyxm | 2 | 0 | 0 | 1 |
| DarthDoger | 2 | 0 | 2 | 0 |
| A-Gentle-Cat | 2 | 0 | 0 | 2 |
| twoxfh | 1 | 0 | 0 | 1 |
| sooxt98 | 1 | 0 | 0 | 0 |
| ShaoxunZeng | 1 | 0 | 1 | 0 |
| Bookraint | 1 | 0 | 0 | 1 |
| arunpatala | 1 | 0 | 0 | 1 |
| dingfangyu | 1 | 0 | 0 | 1 |
| Cyber-Blacat | 1 | 0 | 0 | 1 |
| lucas-vitrus | 1 | 0 | 0 | 0 |
| tasinislam21 | 1 | 0 | 0 | 0 |
| haosenwang1018 | 1 | 0 | 1 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#27dingfangyu2026-03-05 09:07Cannot run the example because of a bfloat16 error in the model_runner.py
- Pull request#26haosenwang10182026-03-03 01:23
- Issue#24yjyddq2026-03-01 03:25Regarding the version compatibility issue between torch, cuda, and flash-attn
- Issue comment#24exlaw2026-02-09 11:38Regarding the version compatibility issue between torch, cuda, and flash-attn
- Issue#24yjyddq2026-02-06 04:58Regarding the version compatibility issue between torch, cuda, and flash-attn
- Issue#23AIxyz2026-01-22 02:48The steps of "Training complete"
- Issue comment#22exlaw2026-01-13 02:48Question about Batch Inference and vLLM Comparison
- Issue comment#21A-Gentle-Cat2026-01-12 13:41Differences between 'generate_wedlm' and 'llm_engine'
- Issue comment#22MatthewCYM2026-01-12 11:21Question about Batch Inference and vLLM Comparison
- Issue#17exlaw2026-01-12 10:55复现一下论文的qwen3-8B-Instruct评测指标的问题
- Issue comment#22exlaw2026-01-12 09:10Question about Batch Inference and vLLM Comparison
- Issue comment#22MatthewCYM2026-01-12 05:58Question about Batch Inference and vLLM Comparison
- Issue comment#21A-Gentle-Cat2026-01-12 05:01Differences between 'generate_wedlm' and 'llm_engine'
- Issue comment#22exlaw2026-01-12 04:00Question about Batch Inference and vLLM Comparison
- Issue#22MatthewCYM2026-01-12 02:25Question about Batch Inference and vLLM Comparison
- Issue comment#20exlaw2026-01-09 09:47The reason of WeDLM-8B-Instruct can be load as Qwen3-8B in vllm
- Issue comment#19exlaw2026-01-08 11:06The infer throughput of WeDLM-8B-Instruct and Qwen3-8B (no_think) in long input
- Issue comment#19AIxyz2026-01-08 08:28The infer throughput of WeDLM-8B-Instruct and Qwen3-8B (no_think) in long input
- Issue comment#19exlaw2026-01-08 07:04The infer throughput of WeDLM-8B-Instruct and Qwen3-8B (no_think) in long input
- Issue#19AIxyz2026-01-08 03:44The infer throughput of WeDLM-8B-Instruct and Qwen3-8B (no_think) in long input
- Issue#18exlaw2026-01-07 01:43Multi thread to curl WeDLM-8B-Instruct server: CUDA error, an illegal memory access was encountered
- Issue comment#18AIxyz2026-01-07 01:03Multi thread to curl WeDLM-8B-Instruct server: CUDA error, an illegal memory access was encountered
- Issue#5exlaw2026-01-06 14:35I created a WeDLM OpenAI-Compatible Server Setup
- Issue comment#18exlaw2026-01-06 14:34Multi thread to curl WeDLM-8B-Instruct server: CUDA error, an illegal memory access was encountered
- Issue#18AIxyz2026-01-06 08:14Multi thread to curl WeDLM-8B-Instruct server: CUDA error, an illegal memory access was encountered
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 260 stars here means stars gained during the window, not the repo's star count.