Demo-131 Cuda graph with optimized paged attention #205
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
PanZezhong1725
requested review from
a team,
Ceng23333,
ma-hang,
pengcheng888,
voltjia,
whjthu,
wooway777 and
zhangyue207
PanZezhong1725
force-pushed
the
demo131
branch
from
be5878b to
4340dff
Compare
Signed-off-by: Ceng23333 <441651826@qq.com>
* issue/204 - support graph in server scripts * issue/208 - adapt to ali ppu * issue/194 - add quantization modify configs accordingly 支持nv w8 1batch 1tp 增加json支持 InfiniLM 增加量化层和global config 以一种比较优雅的方式增加了quant config的支持 修改部分代码结构，删除无用代码 跟随inifnicore修改 删除所有的model_config，统一使用global_config 跟随InfiniLM最新代码修改 修改函数参数顺序 改名global config 为model config Refactor: add new API alongside legacy interfaces with deprecation warnings 添加w4 inifnicore相关内容，以及将Quantization config划入InfiniCore 添加w4 inifnicore相关内容，以及将Quantization config划入InfiniCore * issue/175 - qy device support qy_page_131: add qy device success qy inference_server.py * Issue/170 - Add HYGON support and improve device type handling. * Issue/193: feats for deployment Signed-off-by: Ceng23333 <441651826@qq.com> * skip responding eos token Signed-off-by: Ceng23333 <441651826@qq.com> * issue/143 use add_rmsnorm, nt flash attn, nt kv caching * issue/204 - support graph in server scripts * issue/208 - adapt to ali ppu * rebase main * issue/216 feat: support static kv cache in server * fix llm server cache config * demo131 - resolve mishandled conflicts * demo131 - further adjust attn and caching logic * demo131 - resolve merge requirements --------- Signed-off-by: Ceng23333 <441651826@qq.com> Co-authored-by: wooway777 <wooway777@gmail.com> Co-authored-by: xgqdut2016 <kenan_gewei@163.com> Co-authored-by: gongchensu <zhuyue_134@qq.com> Co-authored-by: Ceng23333 <441651826@qq.com> Co-authored-by: PanZezhong <panzezhong@qiyuanlab.com> Co-authored-by: MaYuhang <2902139028@qq.com>
Issue/221 - resolve cambricon encode plus
issue/219: support vllm bench
Issue/226：add warmup before InfiniLM bench.py generation
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.