Skip to content

Zty dev qwen35 #3

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
benenzhu wants to merge 44 commits into
base: main
Choose a base branch
from
Open

Zty dev qwen35 #3

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
44 commits
Select commit Hold shift + click to select a range
a6b394a
docs(deepeyes): remove missing run_deepeyes_4b.sh refs (#15)
Yangruipis Apr 26, 2026
95f2698
add first qwen3
benenzhu Apr 27, 2026
4b6d22b
zz
benenzhu Apr 27, 2026
23d06ff
feat(amd): add qwen35 rocm smoke
benenzhu Apr 27, 2026
f7102af
merge: sync qwen35 branch with main
benenzhu Apr 27, 2026
d909fef
Update amd/run-qwen35-9b-dapo-math-direct.sh
benenzhu Apr 27, 2026
b5e2a03
Update amd/run_qwen35-9b.sh
benenzhu Apr 27, 2026
f8e7eec
fix(amd): harden qwen35 smoke validation
benenzhu Apr 27, 2026
87221b6
zz
benenzhu Apr 28, 2026
5633f9a
zz
benenzhu Apr 28, 2026
55d61a5
fix: lost multimodal data
NINGBENZHE Apr 27, 2026
3ce5d15
fix(deepeyes): correct judgment spelling
dirtyDan0 Apr 23, 2026
508d939
feat(device): add multi-hardware backend abstraction
NINGBENZHE Apr 27, 2026
d1fbe91
feat(deepeyes): add bbox normalization and Qwen3.5-9B async script
Yangruipis Apr 27, 2026
cc6753e
fix(distributed): add retry and port-release delay for NCCL process g…
Yangruipis Apr 26, 2026
207ace5
fix(dcs): handle Qwen3.5 MoE 2D expert weight format in bridge conver…
Yangruipis Apr 25, 2026
699dd95
zz
benenzhu Apr 28, 2026
0a8e421
add
benenzhu Apr 28, 2026
d22a539
zz
benenzhu Apr 28, 2026
05146b3
add
benenzhu Apr 28, 2026
236c666
zz
benenzhu Apr 28, 2026
58e6758
Merge branch 'zty_dev_qwen35' of https://github.com/AMD-AIM/Relax int…
benenzhu Apr 28, 2026
a12a9e2
zz
Apr 29, 2026
18c1b5e
add
benenzhu Apr 29, 2026
7ee9b58
add
benenzhu Apr 29, 2026
08ccda9
add log then restart machine
benenzhu Apr 29, 2026
50f3639
zz
benenzhu Apr 29, 2026
9f9b771
fix --eps-clip
li126com Apr 29, 2026
e7f70db
add all
benenzhu Apr 30, 2026
b674a4c
Merge remote-tracking branch 'origin/main' into zty_dev_qwen35
benenzhu Apr 30, 2026
71d9a5e
add
benenzhu Apr 30, 2026
b40ca93
feat(rollout): add prefetch, profiling and pipeline encoding
li126com Apr 28, 2026
01973f3
fix(megatron): restore cuda patch target in validate_args
Yangruipis Apr 29, 2026
867d583
chore(data): remove unused prompt helpers
dirtyDan0 Apr 29, 2026
41bcd49
fix(data): avoid prompt mutation in reads
dirtyDan0 May 2, 2026
49bf87b
fix(megatron): wait for previous eval on final training step
NINGBENZHE May 7, 2026
3a25049
feat(megatron): upgrade to Megatron-Bridge mainline
Yangruipis May 7, 2026
5861d8b
feat(models): add Qwen3.6-35B-A3B support with MoE expert detection
Yangruipis May 8, 2026
c1dc570
feat(metrics): report reward dict fields
dirtyDan0 May 9, 2026
2100c15
fix(deepeyes): collapse duplicate image_pad tokens before load_mm_data
Yangruipis May 9, 2026
f18d9b7
Merge remote-tracking branch 'up/main' into zty_dev_qwen35
benenzhu May 13, 2026
8c8b450
zz
benenzhu May 13, 2026
ee5c4cb
zz
benenzhu May 13, 2026
71a949e
zz
benenzhu May 13, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 0 additions & 4 deletions .gitmodules
View file
Open in desktop

This file was deleted.

4 changes: 0 additions & 4 deletions amd/.gitignore
View file
Open in desktop

This file was deleted.

14 changes: 0 additions & 14 deletions amd/README.md
View file
Open in desktop

This file was deleted.

258 changes: 0 additions & 258 deletions amd/qwen3-4b-dapo-math.md
View file
Open in desktop

This file was deleted.

Loading