Skip to content

The most open diffusion language model for code generation — releasing pretraining, evaluation, inference, and checkpoints.

active 2025-09-022026-05-31 (UTC)

Partial coverage11,315 / 12,135 hourly files (93%) · 2 absent upstream · 818 failed, retryable2025-03-232026-08-10 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
337
Pushes
17
Pull requests
3
Issues
18
Stars
250
Forks
16

Activity over time

Daily event counts in the loaded window

Line chart, 272 days from 2025-09-02 to 2026-05-31. Pushes: 17 total, peak 9 in a day. Pull requests: 3 total, peak 2 in a day. Issues: 18 total, peak 2 in a day. Comments: 31 total, peak 6 in a day. Stars: 250 total, peak 31 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

Recent activity

Latest issues, pull requests and releases

  • Issue#31johndpope2026-05-24 08:36
    feat: repr-align improvements — contrastive loss, LLRD, mask curriculum
  • Releasepengzhangzhi2026-03-01 17:35
    published version with DOI
  • Issue#24yuxuanfanOrion2026-01-14 06:24
    Do the code support infer with MacBook with M-series chips?
  • Issue#22AIxyz2025-11-03 09:22
    open-dcoder-0.5B pretrain throughput
  • Issue#18skywalker05232025-10-17 09:37
    some questions about training cost and attention mechanism
  • Issue comment#21pengzhangzhi2025-10-17 02:52
    How many hours to pretrain open-dcoder-0.5B
  • Issue#21AIxyz2025-10-17 02:44
    How many hours to pretrain open-dcoder-0.5B
  • Issue comment#17realgourmet2025-10-06 23:06
    Evaluation Performance Mismatch
  • Issue comment#18skywalker05232025-10-05 04:43
    some questions about training cost and attention mechanism
  • Issue comment#18pengzhangzhi2025-10-04 14:08
    some questions about training cost and attention mechanism
  • Issue comment#1pengzhangzhi2025-10-04 13:17
    Token indices sequence length is longer than the specified maximum sequence length for this model (60022 > 32768). Running this sequence through the model will result in indexing errors
  • Issue comment#1skywalker05232025-10-04 07:29
    Token indices sequence length is longer than the specified maximum sequence length for this model (60022 > 32768). Running this sequence through the model will result in indexing errors
  • Issue comment#17zhangshuibai2025-10-03 16:24
    Evaluation Performance Mismatch
  • Issue comment#10pengzhangzhi2025-10-03 01:49
    RuntimeError: size mismatch during loss weighting in multi-GPU training
  • Issue comment#17pengzhangzhi2025-10-01 01:29
    Evaluation Performance Mismatch
  • Issue#17ZinYY2025-10-01 01:13
    Evaluation Performance Mismatch
  • Issue comment#16pengzhangzhi2025-09-24 05:06
    Code issue
  • Issue#16guoyd19992025-09-24 04:33
    Code issue
  • Issue comment#15pengzhangzhi2025-09-22 15:16
    Input/target length mismatch during pretraining
  • Issue comment#15ZinYY2025-09-22 02:46
    Input/target length mismatch during pretraining
  • Issue#15ZinYY2025-09-22 00:00
    Input/target length mismatch during pretraining
  • Issue#14luonj62025-09-18 07:28
    about rFID
  • Issue comment#13pengzhangzhi2025-09-17 14:33
    Questions about DataCollatorWithPositionIDsMasking sequence concatenation
  • Issue comment#1pengzhangzhi2025-09-17 14:32
    Token indices sequence length is longer than the specified maximum sequence length for this model (60022 > 32768). Running this sequence through the model will result in indexing errors
  • Issue comment#1pharaohcaptain2025-09-17 09:33
    Token indices sequence length is longer than the specified maximum sequence length for this model (60022 > 32768). Running this sequence through the model will result in indexing errors

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 250 stars here means stars gained during the window, not the repo's star count.