| docs: change pr template Co-authored-by: LinShua<707894133@qq.com> # message auto-generated for no-merge-commit merge: !4299 merge master_pr_template3 into master docs: change pr template Created-by: LinShua Commit-by: LinShua Merged-by: ascend-robot Description: ## What this PR does / why we need it? change pr template ## Does this PR introduce any user-facing change? NA ## How was this patch tested? NA See merge request: Ascend/MindSpeed-LLM!4299 | 3 个月前 |
| ci(torch): run st with pytest Co-authored-by: iansheng<shengjiayi@huawei.com> # message auto-generated for no-merge-commit merge: !4403 merge ci412 into master ci(torch): run st with pytest Created-by: iansheng Commit-by: iansheng Merged-by: ascend-robot Description: ## What this PR does / why we need it? Run st test suit with Pytest instead of Shell. ## Does this PR introduce any user-facing change? no ## How was this patch tested? Yes See merge request: Ascend/MindSpeed-LLM!4403 | 2 个月前 |
| feat:Add adaptation support for GLM5.2 model Co-authored-by: LinShua<707894133@qq.com> # message auto-generated for no-merge-commit merge: !4610 merge master_glm5 into master feat:Add adaptation support for GLM5.2 model Created-by: LinShua Commit-by: LinShua Merged-by: ascend-robot Description: ## What this PR does / why we need it? Add adaptation support for GLM5.2 model: 1.Revise and extend the model configuration mapping, add the indexer_types field, and optimize parameter mapping for MoE shared experts. 2.Add dedicated weight conversion scripts for GLM5.2 to support bidirectional checkpoint conversion between HF and Mcore formats, with preconfigured parallelism, MoE and MLA parameters. 3.Implement the share indexer capability, which can be enabled via the arguments --index-topk-freq and --index-skip-topk-offset. 4.Integrate data preprocessing into training scripts and provide supporting environment variable scripts. No breaking changes to external APIs. Internal tests have been completed, with normal inference and precision aligned with vllm-ascend. 5.The MLA function requires enabling the parameters --apply-rope-no-in-complex and --no-use-sparse-c8-indexer. ## Does this PR introduce any user-facing change? NA. ## How was this patch tested? 见PR See merge request: Ascend/MindSpeed-LLM!4610 | 7 天前 |
| fix: Fix Dockerfile and image build script Co-authored-by: wxy1105<wangxinyu136@huawei.com> # message auto-generated for no-merge-commit merge: !4439 merge master into master fix: Fix Dockerfile and image build script Created-by: wxy1105 Commit-by: wxy1105 Merged-by: ascend-robot Description: ## What this PR does / why we need it? Fix Dockerfile and image build script ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## How was this patch tested? Please explain how to verify the correctness and effectiveness of this feature, as well as its usage constraints and limitations. See merge request: Ascend/MindSpeed-LLM!4439 | 1 个月前 |
| docs(pytorch): update checkpoint_convert_hf_mcore.md Co-authored-by: LQ1206<liuqian164@h-partners.com> # message auto-generated for no-merge-commit merge: !4641 merge master into master docs(pytorch): update checkpoint_convert_hf_mcore.md Created-by: LQ1206 Commit-by: LQ1206 Merged-by: ascend-robot Description: ## What this PR does / why we need it? update checkpoint_convert_hf_mcore.md ## Does this PR introduce any user-facing change? Adjust document outline and update decs. ## How was this patch tested? No test See merge request: Ascend/MindSpeed-LLM!4641 | 1 小时前 |
| feat: delete GLM5.2 model data sh Co-authored-by: LinShua<707894133@qq.com> # message auto-generated for no-merge-commit merge: !4627 merge master_glm52_datash into master feat: delete GLM5.2 model data sh Created-by: LinShua Commit-by: LinShua Merged-by: ascend-robot Description: ## What this PR does / why we need it? The glm5.2 already has a script for merging training data, so there is no need for a separate data conversion script, which can be deleted ## Does this PR introduce any user-facing change? NA. ## How was this patch tested? NA. See merge request: Ascend/MindSpeed-LLM!4627 | 2 天前 |
| fix: fix GLM5.2 model hf2mg ckpt conversion Co-authored-by: LinShua<707894133@qq.com> # message auto-generated for no-merge-commit merge: !4635 merge master_glm52_ckf into master fix: fix GLM5.2 model hf2mg ckpt conversion Created-by: LinShua Commit-by: LinShua Merged-by: ascend-robot Description: ## What this PR does / why we need it? GLM5, Deepseek32, and deepseek4 do not have the index_skip_topk_offset and index_topk_freq parameters. Therefore, when these two parameters are absent, it should return true to ensure the normal conversion of other model weights ## Does this PR introduce any user-facing change? No interface changes involved, only environment dependency changes ## How was this patch tested? The UT test case PR will be attached subsequently.. See merge request: Ascend/MindSpeed-LLM!4635 | 2 天前 |
| feat(pytorch): add dsv4 mg2hf Co-authored-by: qyzqyz<quyueze@h-partners.com> # message auto-generated for no-merge-commit merge: !4458 merge master into master feat(pytorch): add dsv4 mg2hf Created-by: qyzqyz Commit-by: qyzqyz Merged-by: ascend-robot Description: ## What this PR does / why we need it? 1. add dsv4 mg2hf - only support pp - only support etp = 1 or tp = 1 2. fix dsv4 hf2mg vpp ## Does this PR introduce any user-facing change? if use base model of dsv4 to do mg2hf convert, please set --model-type-hf with deepseek4_base ## How was this patch tested? Please explain how to verify the correctness and effectiveness of this feature, as well as its usage constraints and limitations. See merge request: Ascend/MindSpeed-LLM!4458 | 1 个月前 |
| fix(pytorch): add swiglu to longcat script Co-authored-by: tichang<tichang@huawei.com> # message auto-generated for no-merge-commit merge: !4636 merge bugfix/add_swiglu_to_longcat_script into master fix(pytorch): add swiglu to longcat script Created-by: tichang Commit-by: tichang Merged-by: ascend-robot Description: ## What this PR does / why we need it? add swiglu to longcat script ## Does this PR introduce any user-facing change? Only the script is modified, and the user plane is not changed. ## How was this patch tested? After adding this parameter, the specified weight can be used to run through the process. See merge request: Ascend/MindSpeed-LLM!4636 | 2 天前 |
| [feat]Modify and improve the configuration of the pre-commit open-source code detection tool Co-authored-by: wujinyuan1<wujinyuan1@huawei.com> # message auto-generated for no-merge-commit merge: !4448 merge master into master [feat]Modify and improve the configuration of the pre-commit open-source code detection tool Created-by: wujinyuan1 Commit-by: wujinyuan1 Merged-by: ascend-robot Description: ## What this PR does / why we need it? pre-commit 工具规则配置修改：试运行阶段发现python bandit工具检测过严，超出昇腾编程规范。 此次修改重点修改bandit规则。 ## Does this PR introduce any user-facing change? 门禁codecheck检测规则精简，提升代码合入效率 ## How was this patch tested? 代码扫描工具配置，不涉及代码仓功能 See merge request: Ascend/MindSpeed-LLM!4448 | 1 个月前 |
| docs(fsdp2): fix documentation format issues and improve readability Co-authored-by: wangjiangben<wangjiangben@huawei.com> # message auto-generated for no-merge-commit merge: !4476 merge docs/fix-fsdp2-docs-format into master docs(fsdp2): fix documentation format issues and improve readability Created-by: wangjiangben Commit-by: wangjiangben Merged-by: ascend-robot Description: ## Summary Fix format errors in FSDP2 documentation and optimize document structure to improve readability and compliance with Markdown standards. **Format Fixes:** - Fix table format issues in quantization.md (missing header separators, column alignment errors) - Fix HTML entity syntax error in arguments.md ("ulysses" missing semicolon) - Remove extra blank lines in code blocks - Standardize indentation in example scripts **Structure Optimization:** - Unify list markers to standard Markdown - syntax - Optimize DTensor section hierarchy for better structure clarity - Convert reference links to proper Markdown link format - Split long paragraphs in quantization descriptions for better readability - Fix MD032 lint error (add blank line before list) **Files Changed:** - arguments.md: Fix HTML entity syntax - fsdp2_basic_features.md: Optimize structure hierarchy and list format - quantization.md: Fix table format, optimize description text See merge request: Ascend/MindSpeed-LLM!4476 | 1 个月前 |
| [feat]Modify and improve the configuration of the pre-commit open-source code detection tool Co-authored-by: wujinyuan1<wujinyuan1@huawei.com> # message auto-generated for no-merge-commit merge: !4448 merge master into master [feat]Modify and improve the configuration of the pre-commit open-source code detection tool Created-by: wujinyuan1 Commit-by: wujinyuan1 Merged-by: ascend-robot Description: ## What this PR does / why we need it? pre-commit 工具规则配置修改：试运行阶段发现python bandit工具检测过严，超出昇腾编程规范。 此次修改重点修改bandit规则。 ## Does this PR introduce any user-facing change? 门禁codecheck检测规则精简，提升代码合入效率 ## How was this patch tested? 代码扫描工具配置，不涉及代码仓功能 See merge request: Ascend/MindSpeed-LLM!4448 | 1 个月前 |
| docs: update README and CONTRIBUTING.md Co-authored-by: xuwenyue<xuwenyue1@huawei.com> # message auto-generated for no-merge-commit merge: !4440 merge master into master docs: update README and CONTRIBUTING.md Created-by: xuwenyue Commit-by: xuwenyue Merged-by: ascend-robot Description: ## What this PR does / why we need it? update README, CONTRIBUTING.md, FAQ.md, features/README.md, performance_tuning.md ## Does this PR introduce any user-facing change? NA ## How was this patch tested? NA See merge request: Ascend/MindSpeed-LLM!4440 | 1 个月前 |
| update LICENSE. add license Signed-off-by: fengliangjun <fengliangjun@huawei.com> | 1 年前 |
| [pytorch][feature] chunkloss for fsdp2 Co-authored-by: sunjunjie1587<sunjunjie8@huawei.com> Co-authored-by: wenjiang23<wenjiang@huawei.com> # message auto-generated for no-merge-commit merge: !3784 merge master into master [pytorch][feature] chunkloss for fsdp2 Created-by: wenjiang23 Commit-by: wenjiang23;sunjunjie1587 Merged-by: ascend-robot Description: 为fsdp2添加chunkloss特性，解决lm_head/cross entropy loss峰值显存问题 See merge request: Ascend/MindSpeed-LLM!3784 | 6 个月前 |
| docs: fix quick_start、README、install_guide docs and imporve user experience Co-authored-by: xuwenyue<xuwenyue1@huawei.com> # message auto-generated for no-merge-commit merge: !4639 merge fix_doc_0624 into master docs: fix quick_start、README、install_guide docs and imporve user experience Created-by: xuwenyue Commit-by: xuwenyue Merged-by: ascend-robot Description: ## What this PR does / why we need it? This PR mainly focuses on improving documentation completeness and user experience for MindSpeed LLM: 1. Added FSDP2 quick start documentation link to README.md and docs_guide.md to provide better navigation for users who want to use the FSDP2 training backend. 2. Added an English placeholder note in docs/en/README.md indicating that documentation translation is in progress, guiding users to refer to Chinese docs for the latest content. 3. Added a table legend explanation in README.md to clarify the meaning of "Released", "✅" and "❌" fields, improving documentation clarity. 4. Added ModelScope download option for model weights in both quick_start.md and fsdp2_quick_start.md, which is more accessible for developers in China who may have difficulty accessing HuggingFace. ## Does this PR introduce any user-facing change? - Documentation updates : Multiple documentation files have been updated with new content and links. - Path changes : - README.md - Added FSDP2 quick start link and table legend - docs/zh/docs_guide.md - Updated FSDP2 documentation link - docs/en/README.md - Added translation in-progress note - docs/zh/pytorch/training/quick_start.md - Added ModelScope weight download method - docs/zh/pytorch/training/fsdp2_quick_start.md - Added ModelScope weight download method ## How was this patch tested? 1. Verified all links are correctly formatted and point to existing documentation files 2. Checked markdown syntax correctness for all changes 3. Confirmed consistency between quick_start.md and fsdp2_quick_start.md for the ModelScope download section 4. Verified that all documentation updates maintain proper structure and readability No code changes were introduced in this PR, so no functional testing is required. The changes only affect documentation content and navigation. See merge request: Ascend/MindSpeed-LLM!4639 | 13 小时前 |
| !2500 update notice Merge pull request !2500 from sunjunjie/master | 1 年前 |
| transformer 4 upgrade 5 & adapt tests Co-authored-by: wanggangguo<wanggangguo@huawei.com> # message auto-generated for no-merge-commit merge: !4545 merge upgrade into master transformer 4 upgrade 5 & adapt tests Created-by: isfrapples Commit-by: wanggangguo Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## How was this patch tested? Please explain how to verify the correctness and effectiveness of this feature, as well as its usage constraints and limitations. See merge request: Ascend/MindSpeed-LLM!4545 | 10 天前 |
| transformer 4 upgrade 5 & adapt tests Co-authored-by: wanggangguo<wanggangguo@huawei.com> # message auto-generated for no-merge-commit merge: !4545 merge upgrade into master transformer 4 upgrade 5 & adapt tests Created-by: isfrapples Commit-by: wanggangguo Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## How was this patch tested? Please explain how to verify the correctness and effectiveness of this feature, as well as its usage constraints and limitations. See merge request: Ascend/MindSpeed-LLM!4545 | 10 天前 |
| refactor(megatron):update coverage script Co-authored-by: guihaowen666<guihaowen@huawei.com> # message auto-generated for no-merge-commit merge: !4295 merge br_master_coverage_fix_0313 into master refactor(megatron):update coverage script Created-by: guihaowen666 Commit-by: guihaowen666 Merged-by: ascend-robot Description: # 覆盖率分析脚本更新 ---- ## What this PR does / why we need it? 更新仓库run_coverage.sh脚本，修复覆盖率分析扫描文件不全的问题 ## Does this PR introduce any user-facing change? 不影响仓库基本功能，旨在优化仓库覆盖率分析功能 ## How was this patch tested? 已在蓝区机器上自测通过 See merge request: Ascend/MindSpeed-LLM!4295 | 3 个月前 |
| refactor(megatron):update coverage script Co-authored-by: guihaowen666<guihaowen@huawei.com> # message auto-generated for no-merge-commit merge: !4295 merge br_master_coverage_fix_0313 into master refactor(megatron):update coverage script Created-by: guihaowen666 Commit-by: guihaowen666 Merged-by: ascend-robot Description: # 覆盖率分析脚本更新 ---- ## What this PR does / why we need it? 更新仓库run_coverage.sh脚本，修复覆盖率分析扫描文件不全的问题 ## Does this PR introduce any user-facing change? 不影响仓库基本功能，旨在优化仓库覆盖率分析功能 ## How was this patch tested? 已在蓝区机器上自测通过 See merge request: Ascend/MindSpeed-LLM!4295 | 3 个月前 |
| feature(pytorch): FSDP2 support hardware-adaptive execution Co-authored-by: zhyebin01<zhangyebin@h-partners.com> # message auto-generated for no-merge-commit merge: !4343 merge fsdp2_gpu into master feature(pytorch): FSDP2 support hardware-adaptive execution Created-by: zhyebin01 Commit-by: zhyebin01 Merged-by: ascend-robot Description: ## What this PR does / why we need it? FSDP2 support hardware-adaptive execution ## Does this PR introduce any user-facing change? No ## How was this patch tested? pipeline test passed See merge request: Ascend/MindSpeed-LLM!4343 | 2 个月前 |
| refactor(megatron):update coverage script Co-authored-by: guihaowen666<guihaowen@huawei.com> # message auto-generated for no-merge-commit merge: !4295 merge br_master_coverage_fix_0313 into master refactor(megatron):update coverage script Created-by: guihaowen666 Commit-by: guihaowen666 Merged-by: ascend-robot Description: # 覆盖率分析脚本更新 ---- ## What this PR does / why we need it? 更新仓库run_coverage.sh脚本，修复覆盖率分析扫描文件不全的问题 ## Does this PR introduce any user-facing change? 不影响仓库基本功能，旨在优化仓库覆盖率分析功能 ## How was this patch tested? 已在蓝区机器上自测通过 See merge request: Ascend/MindSpeed-LLM!4295 | 3 个月前 |
| feat(pytorch): add DeepSeek4 fine-tuning template Co-authored-by: HanhuiChen<chenhanhui1@h-partners.com> # message auto-generated for no-merge-commit merge: !4436 merge dsv4 into master feat(pytorch): add DeepSeek4 fine-tuning template Created-by: HANHU1CHEN Commit-by: HanhuiChen Merged-by: ascend-robot Description: ## What this PR does / why we need it? Adds a fine-tuning template for the DeepSeek4 model series to support its specific prompt format, including thinking mode, tool calling (DSML format), and reasoning effort control. ## Does this PR introduce any user-facing change? Yes — users can now select --prompt-type deepseek4 to fine-tune DeepSeek4 models. Two new behaviors are also exposed: - --enable-thinking controls thinking vs chat mode - --reasoning-effort {max,high} inserts a max-effort instruction prefix; only valid when thinking is enabled - --drop-thinking controls whether reasoning content is kept in each turn ## How was this patch tested? Tested with byte-level alignment against the official encoding_dsv4 script. See merge request: Ascend/MindSpeed-LLM!4436 | 1 个月前 |
| [pytorch][feature]Add RL data processing Co-authored-by: fh_188<fenghui32@huawei.com> # message auto-generated for no-merge-commit merge: !3646 merge fenzhi into master [pytorch][feature]Add RL data processing Created-by: weixin_44917616 Commit-by: fh_188 Merged-by: ascend-robot Description: add preprocess prmompt 修改模板路径 See merge request: Ascend/MindSpeed-LLM!3646 | 7 个月前 |
| feat(pytorch): support o lora rank and q lora rank in v4pro Co-authored-by: dingzicha1997<dingzilin@huawei.com> # message auto-generated for no-merge-commit merge: !4427 merge master into master feat(pytorch): support o lora rank and q lora rank in v4pro Created-by: dingzicha1997 Commit-by: dingzicha1997 Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## How was this patch tested? Please explain how to verify the correctness and effectiveness of this feature, as well as its usage constraints and limitations. See merge request: Ascend/MindSpeed-LLM!4427 | 1 个月前 |
| refactor(megatron):update coverage script Co-authored-by: guihaowen666<guihaowen@huawei.com> # message auto-generated for no-merge-commit merge: !4295 merge br_master_coverage_fix_0313 into master refactor(megatron):update coverage script Created-by: guihaowen666 Commit-by: guihaowen666 Merged-by: ascend-robot Description: # 覆盖率分析脚本更新 ---- ## What this PR does / why we need it? 更新仓库run_coverage.sh脚本，修复覆盖率分析扫描文件不全的问题 ## Does this PR introduce any user-facing change? 不影响仓库基本功能，旨在优化仓库覆盖率分析功能 ## How was this patch tested? 已在蓝区机器上自测通过 See merge request: Ascend/MindSpeed-LLM!4295 | 3 个月前 |
| refactor(megatron):update coverage script Co-authored-by: guihaowen666<guihaowen@huawei.com> # message auto-generated for no-merge-commit merge: !4295 merge br_master_coverage_fix_0313 into master refactor(megatron):update coverage script Created-by: guihaowen666 Commit-by: guihaowen666 Merged-by: ascend-robot Description: # 覆盖率分析脚本更新 ---- ## What this PR does / why we need it? 更新仓库run_coverage.sh脚本，修复覆盖率分析扫描文件不全的问题 ## Does this PR introduce any user-facing change? 不影响仓库基本功能，旨在优化仓库覆盖率分析功能 ## How was this patch tested? 已在蓝区机器上自测通过 See merge request: Ascend/MindSpeed-LLM!4295 | 3 个月前 |
| transformer 4 upgrade 5 & adapt tests Co-authored-by: wanggangguo<wanggangguo@huawei.com> # message auto-generated for no-merge-commit merge: !4545 merge upgrade into master transformer 4 upgrade 5 & adapt tests Created-by: isfrapples Commit-by: wanggangguo Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## How was this patch tested? Please explain how to verify the correctness and effectiveness of this feature, as well as its usage constraints and limitations. See merge request: Ascend/MindSpeed-LLM!4545 | 10 天前 |
| add rl ci Co-authored-by: fh_188<fenghui32@huawei.com> # message auto-generated for no-merge-commit merge: !3951 merge master_rl_ci into master [pytorch][feature]Adapt RL for vllm0110 and add CI Created-by: weixin_44917616 Commit-by: fh_188 Merged-by: ascend-robot Description: [pytorch][feature]Adapt RL for vllm0110 and add CI See merge request: Ascend/MindSpeed-LLM!3951 | 6 个月前 |
| [pytorch][build]update setup.py version and python support range master Co-authored-by: tichang<tichang@huawei.com> # message auto-generated for no-merge-commit merge: !4076 merge chore/update_setup.py_version_and_pyversion_info into master [pytorch][build]update setup.py version and python support range master Created-by: tichang Commit-by: tichang Merged-by: ascend-robot Description: chore(setup): update setup.py version and python support range 在master分支修改了setup.py中的版本信息和python支持版本 See merge request: Ascend/MindSpeed-LLM!4076 | 4 个月前 |
| fix:Fix FSDP2 backend scheduler max step calculation. Co-authored-by: sunjunjie1587<sunjunjie8@huawei.com> # message auto-generated for no-merge-commit merge: !4505 merge master into master fix:Fix FSDP2 backend scheduler max step calculation. Created-by: sunjunjie1587 Commit-by: sunjunjie1587 Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## How was this patch tested? Please explain how to verify the correctness and effectiveness of this feature, as well as its usage constraints and limitations. See merge request: Ascend/MindSpeed-LLM!4505 | 1 个月前 |
