ColossalAI · Issues· 504 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #5037
[DOC]: How to use sgd optimizer?
documentationUpdated Sep 1, 2026 - #5293
[DOC]: Added Japanese README
documentationUpdated Sep 1, 2026 - #5587
[DOC]: What is the datasetset used to train the Colossal-Llama-2?
documentationUpdated Sep 1, 2026 - #6066
[DOC]: 环境安装失败
documentationUpdated Sep 1, 2026 - #6197
[DOC]: Update the documentation of ShardConfig for 1D, 2D, 2.5D, 3D tensor parallelism
documentationUpdated Sep 1, 2026 - #6235
[DOTry to use lora_finetune.py directly and get an error /bin/bash: line 1: export: `=/usr/bin/supervisord': not a valid identifier
documentationUpdated Sep 1, 2026 - #6238
[DOC]: 可以提供使用ColossalAI 训练自定义的模型
documentationUpdated Sep 1, 2026 - #6345
[DOC]: 如何在低版本的CUDA上运行Colossal-AI
documentationUpdated Sep 1, 2026 - #6437
[BUG]: FP8 intranode detection ignores OpenMPI and Slurm local size values
Updated Aug 20, 2026 - #6428
[BUG]: Q/K cross-wiring in `fused_rotary_emb` Triton kernel
bugUpdated Jul 20, 2026 - #6396
[BUG]: `KeyError: 'sum'` When Using Adagrad Optimizer with LowLevelZeroPlugin (initial_scale=8.33)
bugUpdated May 11, 2026 - #6401
[BUG]: FusedLAMB Fails with LowLevelZeroPlugin When Using Small initial_scale and CPU Offload
bugUpdated Apr 14, 2026 - #6387
[BUG]: LowLevelZeroPlugin with use_fp8=true causes `shape mismatch error: shape [32, 512] is invalid for input of size 512`
bugUpdated Apr 14, 2026 - #6413
[BUG]: RuntimeError(f"Could not compute metadata for {kind} target {target}: {e}")
bugUpdated Apr 11, 2026 - #6386
[BUG]: `AssertionError: div_scale should remain default` when using Gemini plugin with CPUAdam
bugUpdated Apr 1, 2026