Skip to content

Pull requests: sgl-project/SpecForge

New pull request New
19 Open 266 Closed
19 Open 266 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Add flash attn backend, duplicates #314
#400 opened Jan 1, 2026 by yubofredwang Loading…
6 tasks
6
[Feature]: Add DeepSeek V3.2 chat template support
#399 opened Jan 1, 2026 by qianlihuang Loading…
1 of 6 tasks
7
feat: support EAGLE-3 speculative decoding training in ling-flash-2.0
#398 opened Dec 29, 2025 by yefei12 Loading…
6 tasks
6
[Clean] Fix warnings and abnormal outputs during training
#393 opened Dec 26, 2025 by lmyybh Loading…
1 of 6 tasks
5
fix data processing + resume training
#390 opened Dec 25, 2025 by Dogacel Loading…
2 of 6 tasks
3
Support check pointing for data generated
#342 opened Dec 1, 2025 by yubofredwang • Draft
6 tasks
feat: added low VRAM flash attention backend
#314 opened Nov 20, 2025 by timmy-feng Loading…
6 tasks done
10
Add Eagle3 training for more MLLM model
#302 opened Nov 15, 2025 by KerwinKai Loading…
6 tasks
14
[Feature] Add accept length simulator for QwenVL
#279 opened Nov 7, 2025 by Lihui-Gu Loading…
6 tasks
6
[Feature]Add Parser for Qwen3 think model
#258 opened Oct 21, 2025 by zyksir Loading…
6 tasks
5
[Feature] Qwen3 VL eagle3 support
#251 opened Oct 10, 2025 by dcw02 Loading…
6 tasks
16
Added mistral model support
#208 opened Sep 1, 2025 by ValeGian Loading…
3 of 6 tasks
21
[Feature] VLM model support tp
#206 opened Sep 1, 2025 by KerwinKai • Draft
6 tasks
3
Support Train Eagle-3 By DeepSpeed
#197 opened Sep 1, 2025 by xq25478 Loading…
13
Adapt Eagle3 for Deepseek architecture
#186 opened Aug 28, 2025 by xuhaojie-2025 Loading…
6 tasks
1
Fix for when draft model hidden dimension is different from target model hidden dimension
#183 opened Aug 26, 2025 by yilian49 Loading…
6
Add Draft LoRA scripts
#138 opened Aug 13, 2025 by shuaills • Draft
6 tasks
11
Feat: Support TP for long-context draft model training high priority
#117 opened Aug 6, 2025 by yd-oom Loading…
6 tasks
7
Added Eagle training support for Kimi-K2
#108 opened Aug 3, 2025 by xuhaojie-2025 Loading…
6
ProTip! Updated in the last three days: updated:>2025-12-29.