Current issue state, recent activity, and per-issue timelines from the indexed issue data.
| Date | Opened | Closed | Comments | Events | Open Backlog |
|---|---|---|---|---|---|
| 2026-07-10 | 0 | 0 | 0 | 0 | 0 |
| 2026-07-09 | 0 | 0 | 0 | 0 | 0 |
| 2026-07-08 | 0 | 0 | 0 | 0 | 0 |
| 2026-07-07 | 0 | 0 | 0 | 0 | 0 |
| 2026-07-06 | 1 | 0 | 0 | 0 | 1 |
| 2026-07-05 | 0 | 0 | 0 | 0 | 0 |
| 2026-07-04 | 0 | 0 | 0 | 0 | 0 |
| 2026-07-03 | 0 | 0 | 0 | 0 | 0 |
| 2026-07-02 | 0 | 0 | 0 | 0 | 0 |
| 2026-07-01 | 0 | 0 | 0 | 0 | 0 |
| 2026-06-30 | 0 | 0 | 0 | 0 | 0 |
| 2026-06-29 | 0 | 0 | 0 | 0 | 0 |
| 2026-06-28 | 0 | 0 | 0 | 0 | 0 |
| 2026-06-27 | 0 | 1 | 0 | 1 | 0 |
Opened: 1
Closed: 0
Comments: 0
Events: 0
| Issue | Author | State | Labels | Comments | Reactions | Updated |
|---|---|---|---|---|---|---|
#6426 [Proposal] HeartFlow - Cognitive Engine for ColossalAI Opened 4 days ago | yun520-1 | open | No labels | 0 | 0 | 4 days ago |
#6249 Does ColossalAI support delayed scaling FP8? Opened 1 year ago | zigzagcai | closed - completed | No labels | 0 | 0 | 13 days ago |
#6396 [BUG]: `KeyError: 'sum'` When Using Adagrad Optimizer with LowLevelZeroPlugin (initial_scale=8.33) Opened 7 months ago | sdjasj | open | bug | 3 | 0 | 2 months ago |
#6401 [BUG]: FusedLAMB Fails with LowLevelZeroPlugin When Using Small initial_scale and CPU Offload Opened 6 months ago | sdjasj | open | bug | 1 | 0 | 3 months ago |
#6387 [BUG]: LowLevelZeroPlugin with use_fp8=true causes `shape mismatch error: shape [32, 512] is invalid for input of size 512` Opened 8 months ago | sdjasj | open | bug | 1 | 0 | 3 months ago |
#6413 [BUG]: RuntimeError(f"Could not compute metadata for {kind} target {target}: {e}") Opened 3 months ago | arminzhu | open | bug | 0 | 0 | 3 months ago |
#6386 [BUG]: `AssertionError: div_scale should remain default` when using Gemini plugin with CPUAdam Opened 8 months ago | sdjasj | open | bug | 1 | 0 | 3 months ago |
#6402 Can you provide the latest Dockerfile? Opened 6 months ago | williamyang4978 | open | No labels | 0 | 0 | 6 months ago |
#6403 Do you have the latest version of Docker images? Opened 6 months ago | williamyang4978 | open | No labels | 0 | 0 | 6 months ago |
#6400 [BUG]: Lamb Optimizer Fails with LowLevelZeroPlugin When Using Small initial_scale and CPU Offload Opened 6 months ago | sdjasj | open | bug | 0 | 0 | 6 months ago |
#6399 [BUG]: RuntimeError When Using GaLoreAdamW8bit with LowLevelZeroPlugin (fp32 + CPU Offload) Opened 6 months ago | sdjasj | open | bug | 0 | 0 | 6 months ago |
#6397 [BUG]: AttributeError: 'PipelineStageManager' object has no attribute 'stage_indices' Opened 6 months ago | honglyua-il | open | bug | 0 | 0 | 6 months ago |
#6389 [BUG]: TypeError in ColoAttention.attention() when enable_all_optimization=True in HybridParallelPlugin Opened 8 months ago | sdjasj | open | bug | 2 | 0 | 7 months ago |
#2938 [BUG]: AttributeError: type object 'Trainer' has no attribute 'add_argparse_args' Opened 3 years ago | pariskang | closed - completed | bug | 21 | 5 | 8 months ago |
#6390 [BUG]: `use_fp8=True` in HybridParallelPlugin causes runtime error `Expected both dimensions of mat2 to be divisible by 16 but got torch.Size([768, 2])` Opened 8 months ago | sdjasj | open | bug | 0 | 0 | 8 months ago |
#6388 [BUG]: GeminiPlugin with `offload_optim_frac=1.0` and small `min_chunk_size_m` causes `RuntimeError: expected input to be on cuda during training` Opened 8 months ago | sdjasj | open | bug | 0 | 0 | 8 months ago |
#6382 [FEATURE]: Add new implementations of RL algorithms. Opened 10 months ago | sglucas | closed - completed | enhancement | 0 | 0 | 10 months ago |
#6381 [BUG]: colossal cannot split tensor evenly when using Sequential Parallelism in hybirdplugin Opened 11 months ago | Hugo-cell111 | open | bug | 2 | 0 | 11 months ago |
#6380 When will ColossalAI support 2D, 2.5D, 3D tensor parallelism? Opened 11 months ago | AmazDeng | open | No labels | 0 | 0 | 11 months ago |
#6379 flux lora train support Opened 11 months ago | AmazDeng | open | No labels | 4 | 0 | 11 months ago |
#6253 [BUG]: can not save model in pipeline training mode Opened 1 year ago | fangxintao | open | bug | 1 | 0 | 11 months ago |
#6358 How to set different parameter learning rate for Hybrid Adam? Opened 1 year ago | crepuscularlight | open | No labels | 2 | 0 | 11 months ago |
#6248 [BUG]: Failed to run lora_finetune.py to fine-tune DS Opened 1 year ago | mahaocong90 | open | bug | 1 | 1 | 11 months ago |
#6367 [FEATURE]: Add more training models and RLHF algorithms Opened 1 year ago | sglucas | open | enhancement | 1 | 0 | 11 months ago |
#6261 [BUG]: Unable to figure out how to pass env variables for each node Opened 1 year ago | Gautam-Rajeev | open | bug | 1 | 0 | 11 months ago |
