Skip to content

Pull requests: HabanaAI/vllm-fork

New pull request New
75 Open 1,931 Closed
75 Open 1,931 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

Optimized hpu_graph of bert models to improve the embedding performance.
#2111 opened Nov 1, 2025 by gyou2021 Loading…
Fix ray init params
#2110 opened Oct 30, 2025 by michalkuligowski Loading…
Improve weights loading and fp8 range conversion on Gaudi2
#2108 opened Oct 30, 2025 by yangulei Loading…
1
Multiple model engine and launch script to support qwen3 reranker
#2106 opened Oct 30, 2025 by tinafengfun Loading…
1
Add support for MiniMax-M2
#2104 opened Oct 29, 2025 by Wei-Lin-Intel Loading…
5
compose small seqlen sdpa for qwen2vl and qwen2.5vl
#2102 opened Oct 29, 2025 by yingjie-han Loading…
3
Fix no attr enable_server_load_tracking error
#2097 opened Oct 28, 2025 by shepark Loading…
Add max_pixels option.
#2094 opened Oct 28, 2025 by wenbinc-Bin Loading…
7
Workaround for Assertion error when embedding with bge-m3 in lazy mode
#2093 opened Oct 28, 2025 by slokesha Loading…
Move only the quantized model and tensors to HPU
#2091 opened Oct 27, 2025 by yangulei Loading…
24
add draft version of vllm inference document for v1.22.0
#2082 opened Oct 24, 2025 by heyuanliu-intel Loading…
3 tasks
Add dotsocr
#2077 opened Oct 23, 2025 by tianyuan211 Loading…
5
fix wrong section for Qwen series doc
#2074 opened Oct 23, 2025 by heyuanliu-intel Loading…
3 tasks
[deepseek_r1]fix weight loading on 1.23
#2073 opened Oct 23, 2025 by ccrhx4 Loading…
1
Enable KEYE_VL on hpu v1.22
#2072 opened Oct 23, 2025 by yingjie-han Loading…
2
Enable chunked prefill on aice 1.22
#2070 opened Oct 23, 2025 by YuJiankang Loading…
refactor(hpu_model_runner): restructure multimodal-related code
#2066 opened Oct 22, 2025 by Jing1Ling • Draft
3 tasks
2
Slokesha port ovis
#2063 opened Oct 21, 2025 by slokesha • Draft
3 tasks
[CS-1549] Eanble function call DeepSeek-V3.1
#2047 opened Oct 19, 2025 by JianyuLi01 Loading…
1
Porting_ovis
#2044 opened Oct 16, 2025 by SupreetSinghPalne • Draft
3 tasks
Spalne/porting ovis
#2038 opened Oct 16, 2025 by SupreetSinghPalne • Draft
3 tasks
replaced apply_rotary_emb_torch() with rotary_embedding imp
#2037 opened Oct 15, 2025 by slokesha • Draft
3 tasks
2
fix bug that VLLM_SKIP_WARMUP=1 is not recognized in vision_bucket
#2036 opened Oct 15, 2025 by yingjie-han Loading…
1
Fix cache miss for Ovis2.5
#2035 opened Oct 15, 2025 by Jianhong-Zhang • Draft
1
Fix cache miss for InternVL
#2034 opened Oct 15, 2025 by Jianhong-Zhang • Draft
ProTip! Add no:assignee to see everything that’s not assigned.