Repository navigation
Commit 282eb1e
Merge branch 'main' into ksivaman/dsv3_4gpu_proxy
Resolve conflicts with the vr200 -> vrnvl72 rename (#6419): rename the
4-GPU DeepSeek V3 proxies to deepseek_v3_pretrain_4gpu_vrnvl72_{fp8mx,nvfp4}_proxy_config
and update their tests and recipe lists.
Follow #6282 and set NVTE_ALLOW_NONDETERMINISTIC_ALGO=0 in both 4-GPU
proxies to match their 256-GPU VRNVL72 parents and the environment invariant test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>636 files changed
Lines changed: 35396 additions & 18800 deletions
File tree
- .github
- ISSUE_TEMPLATE
- actions/test-template
- workflows
- 3rdparty
- docker
- common
- patches
- docs
- _static/model-verification
- fern
- versions
- 0.4.2/pages
- models/llm
- nightly/pages
- modelopt
- models
- bagel
- deepseek
- exaone
- gemma
- glm
- gpt_oss
- kimi
- llama
- minimax
- mistral
- moonlight
- muse_glimmer
- nemotron
- qwen
- stepfun
- training
- modelopt
- models
- bert
- deepseek
- gemma
- glm
- gpt_oss
- llama
- mistral
- nemotron
- qwen
- training
- examples
- conversion
- compare_hf_and_megatron
- megatron_mimo
- llava
- nemotron_omni
- qwen35_vl
- model_verification_cards
- deepseek-v3
- deepseek-v4-flash
- gemma-3-4b-it
- glm5-2
- glm5
- gpt-oss-20b
- nemotron-3-nano-omni-30b-a3b-reasoning
- nemotron-3-super-120b-a12b
- nemotron-3-ultra-550b-a55b
- nemotron-3.5-lightning
- qwen3.6-35b-a3b
- models
- bert/megatron-bert
- deepseek_v4
- diffusion_gemma
- mistral/mistral
- nemotron
- nemotron_3_5_lightning
- nemotron_3_5_super_vl
- iitc/src
- nemotron_3_nano_omni
- nemotron_3_nano
- nemotron_3_super
- nemotron_3_ultra
- nemotron_3
- nemotron_vl
- conf
- training_features/long_context
- scripts
- conversion
- docs
- inference
- performance
- utils
- training
- skills
- adding-model-support
- create-model-verification-card
- scripts
- nemo-mbridge-mlm-bridge-training
- nemo-mbridge-perf-activation-recompute
- nemo-mbridge-perf-cuda-graphs
- nemo-mbridge-perf-parallelism-strategies
- nemo-mbridge-perf-vision-dp-over-cp
- nemo-mbridge-recipe-recommender/references
- src/megatron/bridge
- data
- builders
- collators
- datasets
- energon
- megatron_mimo
- packing
- sources
- diffusion/models/nemotron_labs_diffusion
- models
- bagel
- bert
- common
- conversion
- deepseek
- diffusion_gemma
- exaone/exaone4
- gemma_vl
- gemma
- glm_moe_dsa
- gpt_oss
- hf_pretrained
- kimi
- llama_nemotron
- llama
- megatron_mimo
- conversion
- minimax_m3
- ministral3/data
- mistral
- nemotron_omni
- nemotron_vl
- data
- nemotronh
- nemotron
- qwen3_asr/hf_qwen3_asr
- qwen_vl/modelling_qwen3_vl
- qwen
- sarvam
- peft
- perf_recipes
- deepseek
- gb200
- gb300
- vrnvl72
- glm_moe_dsa
- gb200
- h100
- gpt_oss
- gb300
- vrnvl72
- kimi
- gb300
- vrnvl72
- llama
- vrnvl72
- nemotronh
- b200
- b300
- gb200
- gb300
- h100
- vrnvl72
- qwen_vl
- gb300
- vrnvl72
- qwen
- gb200
- gb300
- vrnvl72
- recipes
- deepseek
- h100
- diffusion_gemma
- gemma
- h100
- glm
- gb200
- h100
- llama
- h100
- moonlight/h100
- nemotron_omni
- nemotron_vl
- h100
- nemotronh_multimodal
- gb200
- h100
- nemotronh
- h100
- qwen_vl
- gb200
- gb300
- h100
- qwen
- gb200
- h100
- utils
- training
- mlm_compat
- post_training/dspark
- utils
- utils
- tests
- functional_tests
- launch_scripts
- gb200/active
- h100/active
- test_groups
- converter
- data/energon
- megatron_mimo
- models
- bert
- diffusion_gemma
- glm_moe_dsa
- qwen
- recipes
- training
- unit_tests
- ci
- conversion/launcher
- data
- builders
- collators
- datasets
- energon
- megatron_mimo
- packing
- sources
- doc_consistency
- examples
- inference
- models
- bert
- common
- deepseek
- diffusion_gemma
- gemma
- glm_moe_dsa
- gpt_oss
- kimi
- llama
- mamba
- megatron_mimo
- conversion
- minimax_m3
- nemotron_omni
- nemotron_vl
- nemotronh
- qwen3_asr
- qwen_vl/modelling_qwen3_vl
- qwen
- sarvam
- peft
- recipes
- nemotronh_multimodal
- nemotronh
- qwen_vl
- qwen
- utils
- scripts
- performance
- training
- skills/create_model_verification_card
- training
- megatron_mimo
- mlm_compat
- post_training
- utils
- utils
- tutorials/recipes/nemotron
Some content is hidden
Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
27 | 27 | | |
28 | 28 | | |
29 | 29 | | |
30 | | - | |
| 30 | + | |
31 | 31 | | |
32 | | - | |
| 32 | + | |
33 | 33 | | |
34 | 34 | | |
35 | 35 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
122 | 122 | | |
123 | 123 | | |
124 | 124 | | |
| 125 | + | |
| 126 | + | |
| 127 | + | |
125 | 128 | | |
126 | 129 | | |
127 | 130 | | |
| |||
177 | 180 | | |
178 | 181 | | |
179 | 182 | | |
| 183 | + | |
| 184 | + | |
| 185 | + | |
180 | 186 | | |
181 | 187 | | |
182 | 188 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
27 | 27 | | |
28 | 28 | | |
29 | 29 | | |
| 30 | + | |
| 31 | + | |
| 32 | + | |
30 | 33 | | |
31 | 34 | | |
32 | 35 | | |
| |||
This file was deleted.
This file was deleted.
0 commit comments