Skip to content

Pull requests: HabanaAI/vllm-fork

New pull request New
46 Open 557 Closed
46 Open 557 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Enable splitting qkv and gate_up
#662 opened Dec 24, 2024 by tianmu-li Loading…
[SW-197036] - use torch._scaled_mm with hpu
#660 opened Dec 22, 2024 by nirda7 Loading…
1
Draft: Delayed prompts
#659 opened Dec 20, 2024 by kamil-kaczor • Draft
Port "Fix recompilations due to different batch_sizes in MSS PR637"
#657 opened Dec 20, 2024 by iboiko-habana Loading…
Chunked Prefill
#656 opened Dec 20, 2024 by hlahkar • Draft
Lora manager tests fix
#652 opened Dec 19, 2024 by rsshaik1 • Draft
Add mark_step for encoder layers
#650 opened Dec 19, 2024 by yisonzhu Loading…
1
[bugfix] fix RuntimeError on apc
#648 opened Dec 19, 2024 by kkimmk Loading…
Support throughput benchmarking for mllama with vision input
#647 opened Dec 19, 2024 by yisonzhu Loading…
Fix: selecting correct backend for MultiHeadAttention habana Issues or PRs submitted by Habana Labs
#645 opened Dec 18, 2024 by adobrzyniewicz-habana Loading…
Fix model OOM issue in llama-405 and mixtral - 2nd attempt habana Issues or PRs submitted by Habana Labs
#644 opened Dec 18, 2024 by afierka-intel Loading…
@afierka-intel
Selective merged prefill
#643 opened Dec 18, 2024 by xuechendi Loading…
Add exponential bucketing integration
#642 opened Dec 17, 2024 by kzawora-intel • Draft
Multimodality fix for llava habana Issues or PRs submitted by Habana Labs
#641 opened Dec 17, 2024 by adobrzyniewicz-habana Loading…
Add inc fp8 qunatization documentation
#635 opened Dec 16, 2024 by nirda7 Loading…
47
Device Type HPU support for torch.generator() API
#628 opened Dec 13, 2024 by nageshdn Loading…
1
Fix long contexts in LoRA
#624 opened Dec 12, 2024 by SanjuCSudhakaran Loading…
Use FusedSDPA for MllamaVisionSdpaAttention
#620 opened Dec 11, 2024 by kdamaszk • Draft
[BUG fix] Rebase caused spec decode fix
#613 opened Dec 11, 2024 by xuechendi Loading…
11
[WIP] Add HPU support to vLLM v1 - cont.
#609 opened Dec 10, 2024 by kzawora-intel Loading…
21 of 23 tasks
Optimize for topk=1 case if we do not handle duplicates
#603 opened Dec 9, 2024 by ssarkar2 Loading…
2
Add in Dockerfile.hpu.ubi
#602 opened Dec 9, 2024 by Xaenalt Loading…
13
Add real BS & seq_len to profiling
#601 opened Dec 9, 2024 by kamil-kaczor Loading…
Documentation update for 1.19
#597 opened Dec 5, 2024 by PatrykWo Loading…
20
[DONNOT MERGE] [POC] Ray Compiled graphs
#593 opened Dec 5, 2024 by michalkuligowski • Draft
ProTip! Add no:assignee to see everything that’s not assigned.