xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-01-07 00:03:07 +08:00
Code Issues Packages Projects Releases Wiki Activity
4,395 Commits 155 Branches 89 Tags
Commit Graph

15 Commits

Author SHA1 Message Date
Chen Zhang
89003c4082
[v1][Bugfix] Add extra_keys to block_hash for prefix caching (#12603) 
This pr adds extra key to block hash, to generate different hash value
for two blocks with the same token string but different extra_keys in
their parent blocks. For example, it can generate different hash value
for the second block of the following two requests:
```python
request1 = make_request(
        request_id=0,
        prompt_token_ids=[_ for _ in range(6)],
        mm_positions=[{
            "offset": 0,
            "length": 3
        }, {
            "offset": 3,
            "length": 3
        }],
        mm_hashes=["hash1", "hash2"],
    )
    request2 = make_request(
        request_id=1,
        prompt_token_ids=[_ for _ in range(6)],
        mm_positions=[{
            "offset": 0,
            "length": 3
        }, {
            "offset": 3,
            "length": 3
        }],
        mm_hashes=["hash3", "hash2"],
    )
```

---------

Signed-off-by: Chen Zhang <zhangch99@outlook.com>
 2025-01-31 13:13:04 -08:00
Cody Yu
f0ef37233e
[V1] Add uncache_blocks (#12333) 2025-01-23 04:19:21 +00:00
Cody Yu
7206ce4ce1
[Core] Support reset_prefix_cache (#12284) 2025-01-22 18:52:27 +00:00
Chen Zhang
994fc655b7
[V1][Prefix Cache] Move the logic of num_computed_tokens into KVCacheManager (#12003) 2025-01-15 07:55:30 +00:00
Roger Wang
91b361ae89
[V1] Extend beyond image modality and support mixed-modality inference with Llava-OneVision (#11685) 
Signed-off-by: Roger Wang <ywang@roblox.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2025-01-06 19:58:16 +00:00
Jee Jee Li
47831430cc
[Bugfix][V1] Fix test_kv_cache_utils.py (#11738) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-01-04 16:07:59 +00:00
xcnick
d91457d529
[V1] Add kv cache utils tests. (#11513) 
Signed-off-by: xcnick <xcnick0412@gmail.com>
 2025-01-04 14:49:46 +08:00
Chen Zhang
8c3230d8c1
[V1] Simpify vision block hash for prefix caching by removing offset from hash (#11646) 2024-12-31 08:56:01 +00:00
sakunkun
2c5718809b
[Bugfix] Move the _touch(computed_blocks) call in the allocate_slots method to after the check for allocating new blocks. (#11565) 2024-12-31 06:29:04 +00:00
Cody Yu
bf8717ebae
[V1] Prefix caching for vision language models (#11187) 
Signed-off-by: Cody Yu <hao.yu.cody@gmail.com>
 2024-12-17 16:37:59 -08:00
Cody Yu
78ed8f57d8
[Misc][V1] Fix type in v1 prefix caching (#11151) 2024-12-13 00:57:40 +00:00
Woosuk Kwon
a79b122400
[V1] Do not allocate beyond the max_model_len (#10730) 
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
 2024-11-28 00:13:15 -08:00
Ricky Xu
97814fbf0f
[v1] Refactor KVCacheManager for more hash input than token ids (#10507) 
Signed-off-by: rickyx <rickyx@anyscale.com>
Signed-off-by: Cody Yu <hao.yu.cody@gmail.com>
Co-authored-by: Cody Yu <hao.yu.cody@gmail.com>
 2024-11-22 23:27:25 +00:00
Cyrus Leung
0b8bb86bf1
[1/N] Initial prototype for multi-modal processor (#10044) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2024-11-13 12:39:03 +00:00
Cody Yu
201fc07730
[V1] Prefix caching (take 2) (#9972) 
Signed-off-by: Cody Yu <hao.yu.cody@gmail.com>
 2024-11-07 17:34:44 -08:00