-
Notifications
You must be signed in to change notification settings - Fork 4.3k
Insights: hpcaitech/ColossalAI
Overview
-
- 17 Merged pull requests
- 7 Open pull requests
- 4 Closed issues
- 0 New issues
Could not load contribution data
Please try again later
17 Pull requests merged by 9 people
-
[shardformer] fix import
#5788 merged
Jun 6, 2024 -
[misc] update requirements
#5787 merged
Jun 6, 2024 -
[install]fix setup
#5786 merged
Jun 6, 2024 -
[misc] fix dist logger
#5782 merged
Jun 5, 2024 -
Allow building cuda extension without a device.
#5535 merged
Jun 5, 2024 -
[gemini] optimize reduce scatter d2h copy
#5760 merged
Jun 5, 2024 -
[hotfix] fix testcase in test_fx/test_tracer
#5779 merged
Jun 5, 2024 -
[Test/CI] remove test cases to reduce CI duration
#5753 merged
Jun 5, 2024 -
[misc] Accelerate CI for zero and dist optim
#5758 merged
Jun 5, 2024 -
[hotfix] fix llama flash attention forward.
#5777 merged
Jun 5, 2024 -
[Inference]Add Streaming LLM
#5745 merged
Jun 5, 2024 -
[devops] fix docker ci
#5780 merged
Jun 4, 2024 -
[misc] update dockerfile
#5776 merged
Jun 4, 2024 -
[CI/tests] simplify some test case to reduce testing time
#5755 merged
Jun 4, 2024 -
[Hotfix] Add missing init file in inference.executor
#5774 merged
Jun 3, 2024 -
Fix/fix testcase
#5770 merged
Jun 3, 2024 -
[shardformre] fix llama policy
#5765 merged
Jun 3, 2024
7 Pull requests opened by 6 people
-
[Inference] Refactor inference modeling
#5771 opened
Jun 3, 2024 -
Multi round dialogue branch
#5772 opened
Jun 3, 2024 -
[Feauture] MoE refactor
#5775 opened
Jun 4, 2024 -
[Gemini] Use async stream to prefetch and h2d data moving
#5781 opened
Jun 5, 2024 -
[Inference]Lazy Init Support
#5785 opened
Jun 6, 2024 -
Support 4D parallel + Flash Attention
#5789 opened
Jun 6, 2024 -
fix Llama rotary embedding api change for transformers 4.39.3
#5790 opened
Jun 7, 2024
4 Issues closed by 2 people
-
[PROPOSAL]: Refactor inference engine by selecting backend during init of modules
#5773 closed
Jun 7, 2024 -
[BUG]: Cannot build extensions when no gpu device exists
#5534 closed
Jun 5, 2024 -
[BUG]:Report some errors in test_fx/test_tracer
#5778 closed
Jun 5, 2024 -
[BUG]: Report some errors in test
#5768 closed
Jun 3, 2024
7 Unresolved conversations
Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.
-
[Feature] optimize PP overlap
#5735 commented on
Jun 7, 2024 • 12 new comments -
[ColossalChat] Colossalchat upgrade
#5759 commented on
Jun 7, 2024 • 6 new comments -
[FEATURE]: SP with FlashAttention
#5762 commented on
Jun 3, 2024 • 1 new comment -
[BUG]: docker build cuda extension error
#5732 commented on
Jun 6, 2024 • 1 new comment -
[lora] support lora for Gemini
#5001 commented on
Jun 1, 2024 • 1 new comment -
[WIP][Infer] Inference Distributed RPC Framework Optimization
#5756 commented on
Jun 5, 2024 • 0 new comments -
[moe/zero] refactor low level optimizer
#5767 commented on
Jun 7, 2024 • 0 new comments